A two-strain starter culture containing Lactobacillus plantarum CIDCA 83114, a potential probiotic strain isolated from kefir grains, and Streptococcus thermophilus CIDCA 321 was tested for the preparation of a fermented milk product. Kluyveromyces marxianus CIDCA 8154, a yeast with immunomodulatory properties was included to formulate a three-strain starter culture. Supernatants of enterohaemorragic Escherichia coli, shiga-toxin–producing strain, along with a two-strain or a three-strain starter culture were included in the medium of Vero-cell surface cultures. The results demonstrated that these combinations of microorganisms antagonize the cytopathic action of shiga toxins. The cell concentration of Lb. plantarum did not decrease during fermentation, indicating that the viability of this strain was not affected by low pH, nor did the number of viable bacteria change during 21 days of storage in either fermented products. The number of viable yeasts increases during fermentation and storage. Trained assessors analyzed the general acceptability of fresh fermented milks and considered both acceptable. The milk fermented with the two-strain starter culture was considered acceptable after two week of storage, while the product fermented with the three-strain starter culture remained acceptable for less than one week. The main changes in sensory attributes detected by the trained panel were in sour taste, milky taste and also in fermented attributes. The correlation between different sensory attributes and acceptability indicated that the panel was positively influenced by milky attributes (taste, odour, and flavour) as well as the intensity of flavour. In conclusion, the two-strain starter culture would be the more promising alternative for inclusion of that potential probiotic lactobacillus in a fermented milk product.

The aim of this study was to evaluate the antioxidant, antihypertensive and immunomodulatory characteristics of skim milk fermented with Lactobacillus delbrueckii ssp. bulgaricus LB340. Supernatants obtained from the ferments after centrifugation were subjected to ultrafiltration and yielded four peptidic fractions of 10–5 kDa, 5–3 kDa, 3–1 kDa, and <1·0 kDa. Peptides in 5–3 kDa range exhibited a good antioxidant activity. The peptides (<1·0 k) was applied to Superdex-30 G column fractionation and produced six fractions (F1–6). Fraction F2 presented the highest angiotensin I-converting enzyme inhibition activity with IC50 of 67·71±7·62 mg/ml. Moreover, fraction F6, which displayed a good immunomodulatory activity, had a positive effect on murine spleen lymphocyte proliferation with Stimulation Index of 0·729±0·123. The present data showed the potential of the milk fermented with Lactobacillus delbrueckii ssp. bulgaricus LB340 as a functional food, however, further research is needed to evaluate the biofunctional activity of this fermentation product in vivo using model animal.

This work studied the behaviour of caseinomacropeptide (CMP) in a whey protein fractionation process based on the selective precipitation of α-lactalbumin (α-la) in an acid medium. Three different acids (hydrochloric, citric and lactic) and different operating conditions (protein concentration, temperature and pH) were considered to perform the precipitation step. Under the optimised precipitation conditions obtained for α-la (pH 4, 55°C, initial α-la concentration around 12 g/l) CMP presents quite similar behaviour to that observed for β-lactoglobulin (β-lg), namely remaining in the supernatant fraction. However, at a lower pH value (3·5) the amount of precipitated CMP increases up to 72% when citric acid is added. This behaviour could be due to the fact that CMP is close to its isoelectric point, which allows a supernatant fraction enriched in β-lg that is almost free from the rest of proteins in sweet whey.

Stereospecific analysis is an important tool for the characterization of lipid fraction of food matrices, and also of milk samples. The results of a chemical-enzymatic-chromatographic analytical method were elaborated by chemometric procedures such as linear discriminant analysis (LDA) and artificial neural network (ANN). According to the total composition and intrapositional fatty acid distribution in the triacylglycerol (TAG) backbone, the obtained results were able to characterize pure milk samples and milk mixtures with 1, 3, 5% cow milk added to donkey milk. The resulting score was very satisfactory. Totally correct classified samples were obtained when the TAG stereospecific results of all the considered milk mixtures (donkey-cow) were elaborated by LDA and ANN chemometric procedures.

The effect of freezing on the properties of a raw ewes’-milk semi-soft cheese (Serpa cheese) was studied using small amplitude oscillatory (SAOS) and texture measurements, colour and chemical parameters. The freezing was introduced at three different stages of the ripening process (28, 35 and 42 days), and the cheeses were maintained frozen for 12 months. Cheeses were submitted to a slow or fast freezing method, and to different storage temperatures: −10 and −20°C (three replicates for each set conditions). Chemical data showed that only the proteolysis indicators exhibited differences between frozen and non-frozen samples; frozen samples showed higher values of NPN than the non-frozen samples, indicating that the freezing process did not prevent the secondary proteolysis of cheese. Frozen samples showed a significantly (P<0·05) stronger structure than the non-frozen, as indicated by hardness. However, the differences between the frozen and non-frozen samples were not significantly for storage modulus (G′1Hz) and loss tangent (tan δ1Hz) (P>0·05). Freezing affected mainly colour parameters: frozen samples were more luminous, and more yellow-green. The results allowed us to conclude that the damages caused by freezing to cheese properties could be minimized if this type of storage is introduced at the end of ripening (42 d) using a freezing temperature of −20°C.

Proteinase activity of Lactobacillus fermentum IFO3956 cells was higher when they were grown on milk-based media than on 10% reconstituted skim milk. The lowest protease activity was observed when cells were grown on milk-free media. The extraction of milk-induced cell-bound proteases from Lb. fermentum IFO3956 was most efficient using 1% Tween 80 while the use of 1% SDS inhibited all proteolytic activity. Two bands of ∼35 and >100 kDa were observed by zymogram, indicating that proteolytic activity corresponded to the presence of at least two types of enzymes or two molecular forms of one enzyme. Mass spectrometry analyses of αS1-casein hydrolysates detected 24 peptides with sizes ranging from 5 to 36 amino acids, including 9 phosphorylated peptides, resulting from the fermentation of Lb. fermentum IFO3956 of αS1-casein. Most of the identified peptides originated from the N-terminal portion of αS1-casein. The studied bacterial strain could hydrolyze αS1-casein in many sites including the epitopes triggering the allergic reactions against αS1-casein e.g. at the positions 23, 30, 41, 71, 91, 98, 126, 179. After hydrolysis of αS1-casein with Lb. fermentum IFO3956 the recognition and the binding of this casein to IgE from the pooled sera of 18 patients with cow's milk allergy was significantly reduced.

This study evaluated the effect of storage on renneting properties of goat milk investigated using the Formagraph method. Milk samples from 169 goats in three farms (F1, F2 and F3) were analysed during an entire lactation (45, 75, 105, 135 and 165 days in milking DIM), to obtain renneting parameters, both from fresh milk and after storage with Bronopol and freezing at −20°C and −80°C. As regards fresh milk, mean values of clotting time were between 12·51 (45 DIM) and 13·29 min (105 DIM and F2), the curd firming time between 1·77 (45 DIM) and 2·15 min (F1) and curd firmness between 42·09 (165 DIM) and 49·55 mm (45 DIM). No statistical difference was recorded after storage. After regression analysis, all prediction models showed significance value at P<0·001 with the highest R2 value for clotting time, 0·710 (fresh vs. frozen milk at −20°C), and the lowest for clot firmness, 0·281 (fresh vs. frozen milk at −80°C). Results demonstrated that assessment of goat milk coagulation properties using the Formagraph method is also achievable after freezing or Bronopol addition.

The objectives of the study were to compare the ovarian activity of Holstein-Friesian (CH HF), Fleckvieh (CH FV) and Brown Swiss (CH BS) dairy cows of Swiss origin with that of Holstein-Friesian (NZ HF) dairy cows of New Zealand origin, the latter being used as a reference for reproductive performance in pasture-based seasonal calving systems. Fifty, second-lactation NZ HF cows were each paired with a second-lactation Swiss cow (17, 15 and 18 CH HF, CH FV and CH BS respectively) in 13 pasture-based, seasonal-calving commercial dairy farms in Switzerland. Ovarian activity was monitored by progesterone profiling from calving to first breeding service. CH BS cows produced less energy-corrected milk (mean 22·8 kg/d) than the other breeds (26·0–26·5 kg/d) during the first 100 d of lactation. CH HF cows had the lowest body condition score (BCS) at calving and the greatest BCS loss from calving to 30 d post partum. Commencement of luteal activity (CLA) was later for NZ HF than for CH FV (51·5 v. 29·2 d; P <0·01), with CH HF and CH BS intermediate (43 d). On average, NZ HF and CH HF cows had one oestrous cycle before the onset of the seasonal breeding period; this was less (P<0·01) than either CH FV (1·7) or CH BS (1·6). There was a low prevalence of luteal persistency (3%) among the studied cows. First and second oestrous cycle inter-ovulatory intervals did not differ between breeds (20·5–22·6 d). The luteal phase length of CH BS during the second cycle was shorter (10·6 d) than that of the other breeds (13·8–16·0 d), but the inter-luteal interval was longer (9·8 d v. 7·0–8·0 d). The results suggest that the Swiss breeds investigated have a shorter interval from calving to CLA than NZ HF cows.

The aim of this study was to investigate the effects of enzymatic hydrolysis with digestive enzymes of camel whole casein and beta-casein (β-CN) on their antioxidant and Angiotensin Converting Enzyme (ACE)-inhibitory properties. Peptides in each hydrolysate were fractionated with ultra-filtration membranes. The antioxidant activity was determined using a Trolox equivalent antioxidant capacity (TEAC) scale. After enzymatic hydrolysis, both antioxidant and ACE-inhibitory activities of camel whole casein and camel β-CN were enhanced. Camel whole casein and β-CN showed significant ACE-inhibitory activities after hydrolysis with pepsin alone and after pepsinolysis followed by trypsinolysis and chymotrypsinolysis. Camel β-CN showed high antioxidant activity after hydrolysis with chymotrypsin. The results of this study suggest that when camel milk is consumed and digested, the produced peptides start to act as natural antioxidants and ACE-inhibitors.

The effect of ultrafiltration and diafiltration on the distribution of the calcium phosphate clusters of the casein micelle was investigated using Small Angle Neutron Scattering (SANS). In the case of ultrafiltration, fresh skim milk was subjected to concentration using membrane filtration up to 5× its original volume, the retentate was rediluted with its corresponding serum and subsequently dialyzed against reconstituted milk powder dispersed in D2O/H2O (UF 5×(D)). In the case of diafiltered samples, the samples were concentrated adding water (diafiltration) at two different levels (DF 2·5× or DF 5×) and then redispersed with D2O/H2O. In the DF 5× case, the serum components were diluted to less than 1% of their original concentration. For analysis, all samples had the same volume fraction of dispersed casein micelles (φ=0·1), which is that of the control, unprocessed skim milk. A peak in the SANS data was observed in fresh skim milk at a scattering vector, qo, of 0·034 Å−1 (directly proportional to the reciprocal characteristic length), in agreement with previous literature results. Neutron data on the ultrafiltered, UF 5×(D) and diafiltered, DF 2·5× and DF 5× milk samples showed a progressive decrease in the intensity of the peak but invariance in qo. These results, combined with the determination of soluble and insoluble calcium in the samples, suggest a progressive and irreversible removal of calcium from within the micelle during membrane filtration of milk. Using SANS it was possible to clearly show changes in the micellar calcium clusters that may not otherwise be fully determined by only measuring the amount of total and insoluble calcium in milk.

Dairy herds are expanding and, with increasing numbers of animals in each herd, there is a need for automatic recording of indicators in milk in order to detect mastitis, inflammation of the udder. A number of biomarkers for mastitis have been suggested over the years. Mastitis usually occurs in one of the four udder quarters and since it is now possible to milk each udder quarter separately in automated milking systems, it is important to evaluate the normal variation in the biomarkers at udder quarter level. This study evaluated the normal variations between milkings for some biomarkers in clinically healthy cows, determined by repeated somatic cell count and bacteriological analysis. The biomarkers studied were serum amyloid A (SAA), haptoglobin (Hp), lactate dehydrogenase (LDH), N-acetyl-β-d-glucosaminidase (NAGase) and alkaline phosphatase (AP), parameters that have been suggested as markers for mastitis. Ten cows were monitored on 42 consecutive milking occasions through collection of udder quarter milk samples and representative cow composite milk samples, giving a total of 2100 individual milk samples. Each cow had its individual profile for the concentrations and variations in the parameters analysed. Although there was relatively large variation between cows for the biomarkers analysed, the variation between milkings in clinically healthy quarters within cows was often below 10%. The biomarker with the lowest variation in this study was LDH. The results suggest that comparing quarters within an individual cow can identify deviations from the natural variations between milkings. This could be a valuable tool instead of, or in combination with, a cut-off value for each parameter in order to detect changes in the milk indicating mastitis.

The aim of the study was to describe the characteristics of cleaning procedures for milking equipment applied in intensive dairy farms in Lombardy (Italy) and to study their relationships with bacterial count of bulk milk and hygienic condition of milking machine components. A group of 22 dairy farms was visited twice (winter and summer) in order to collect bulk tank milk and post-rinse water samples and swabs from liners and milk receiver. Samples were analysed to determine: standard plate count (SPC), laboratory pasteurization count (LPC), psychrotrophic bacteria count (PBC), coliform count (CC) and Escherichia coli. Cleaning procedures were monitored using electronic milk flow meters with specific software for the measurement of the duration of each cleaning phase, circulating solution temperature and electrical conductivity, turbulence and water filling percentage of pipelines. The results showed that farms classified as high and low milk total bacteria count significantly differed both in terms of liners and receiver bacterial contamination and in terms of water temperature reached during the detergent phase of cleaning milking equipment. Significant positive correlations were found among total bacteria count in milk and bacterial contamination of the liners. Maximum water temperature reached during the cleaning cycle of milking equipment was very low (34·4±8·9°C on average); most of the observations (88·6%) corresponded to water temperatures <45°C. Cleaning temperature was related to psychrotrophic bacteria count of milk and post-rinse water and coliform count in liners. Routine check and regulation of water temperature during the washing phase of the milking machine can be a simple and effective way to control one of the main risk factors for bacteriological quality of bulk tank milk.

The capacity of lactic acid bacteria to produce exopolysaccharides (EPS) conferring microorganisms a ropy phenotype could be an interesting feature from a technological point of view. Progressive adaptation to bile salts might render some lactobacilli able to overcome physiological gut barriers but could also modify functional properties of the strain, including the production of EPS. In this work some technological properties and the survival ability in simulated gastrointestinal conditions of Lactobacillus delbrueckii subsp. lactis 193, and Lb. delbrueckii subsp. lactis 193+, a strain with stable bile-resistant phenotype derived thereof, were characterized in milk in order to know whether the acquisition of resistance to bile could modify some characteristics of the microorganism. Both strains were able to grow and acidify milk similarly; however the production of ethanol increased at the expense of the aroma compound acetaldehyde in milk fermented by the strain 193+, with respect to milk fermented by the strain 193. Both microorganisms produced a heteropolysaccharide composed of glucose and galactose, and were able to increase the viscosity of fermented milks. In spite of the higher production yield of EPS by the bile-resistant strain 193+, it displayed a lower ability to increase viscosity than Lb. delbrueckii subsp. lactis 193. Milk increased survival in simulated gastric juice; the presence of bile improved adhesion to the intestinal cell line HT29-MTX in both strains. However, the acquisition of a stable resistance phenotype did not improve survival in simulated gastric and intestinal conditions or the adhesion to the intestinal cell line HT29-MTX. Thus, Lb. delbrueckii subsp. lactis 193 presents suitable technological properties for the manufacture of fermented dairy products; the acquisition of a stable bile-resistant phenotype modified some properties of the microorganism. This suggests that the possible use of bile-resistant derivative strains should be carefully evaluated in each specific application considering the influence that the acquisition of a stable bile-resistant phenotype could have in survival ability in gastric and intestinal conditions and in technological properties.

Milk fatty acid (FA) profile is a dynamic pattern influenced by lactational stage, energy balance and dietary composition. In the first part of this study, effects of the energy balance during the proceeding lactation [weeks 1–21 post partum (pp)] on milk FA profile of 30 dairy cows were evaluated under a constant feeding regimen. In the second part, effects of a negative energy balance (NEB) induced by feed restriction on milk FA profile were studied in 40 multiparous dairy cows (20 feed-restricted and 20 control). Feed restriction (energy balance of −63 MJ NEL/d, restriction of 49 % of energy requirements) lasted 3 weeks starting at around 100 days in milk. Milk FA profile changed markedly from week 1 pp up to week 12 pp and remained unchanged thereafter. The proportion of saturated FA (predominantly 10:0, 12:0, 14:0 and 16:0) increased from week 1 pp up to week 12 pp, whereas monounsaturated FA, predominantly the proportion of 18:1,9c decreased as NEB in early lactation became less severe. During the induced NEB, milk FA profile showed a similarly directed pattern as during the NEB in early lactation, although changes were less marked for most FA. Milk FA composition changed rapidly within one week after initiation of feed restriction and tended to adjust to the initial composition despite maintenance of a high NEB. C18:1,9c was increased significantly during the induced NEB indicating mobilization of a considerable amount of adipose tissue. Besides 18:1,9c, changes in saturated FA, monounsaturated FA, de-novo synthesized and preformed FA (sum of FA >C16) reflected energy status in dairy cows and indicated the NEB in early lactation as well as the induced NEB by feed restriction.

This work aimed to test if milk preserved with bronopol can be reliably used for fatty acid determination. Dairy production and milk quality are often monitored regularly to assess performance and contribute to selection indices. With evidence that fat composition can be influenced by selective breeding, there might be an interest in using samples collected in routine testing to evaluate individual cow fatty acid profiles, contributing to breeding indices. However, most recording services use a preservative such as bronopol and there is no published record if this influences subsequent fatty acid analysis. This study used milk from an oil seed supplementation trial, generating a wide range of milk fatty acid profiles, to test if the concentration of 31 individual fatty acids determined by GC were influenced by bronopol. Provided preserved samples are subsequently frozen, milk treated with bronopol can reliably be used to evaluate fatty acid composition in most cases; however bronopol might influence a few long-chain fatty acids present in relatively low concentrations. This is one small step towards simplifying milk compositional analysis but it could ultimately streamline the inclusion of milk fat quality into breeding indices, either with a view to ‘healthier’ milk or potentially reducing methane output and the environmental impact of dairy production.

Experiments were designed to investigate the suitability of a combination of a short manual teat stimulation with a short latency period before teat cup attachment to induce and maintain oxytocin release and milk ejection without interruption. In Experiment 1, seven dairy cows in mid lactation were manually pre-stimulated for 15, 30 or 45 s, followed by either 30 s or 45 s of latency period. It was shown that all treatments induced a similar release of oxytocin without interruption until the end of milking. In particular, the latency period of up to 45 s did not cause a transient decrease of oxytocin concentration. In Experiment 2, milking characteristics were recorded in seven cows each in early, mid, and late lactation, respectively. Because the course of milk ejection depends mainly on the degree of udder filling, individual milkings were classified based on the actual degree of udder filling which differs between lactational stages but also between morning and evening milkings. All animals underwent twelve different udder preparation treatments, i.e. 15, 30, or 45 s of pre-stimulation followed by latency periods of 0, 30, 45, or 60 s. Milking characteristics were recorded. Total milk yield, main milking time and average milk flow rate did not differ between treatments if the degree of udder filling at the start of milking was >40% of the maximum storage capacity. However, if the udder filling was <40%, main milking time was decreased with the duration of a latency period up to 45 s, independent of duration of pre-stimulation. Average milk flow at an udder filling of <40% was highest after a pre-stimulation of 45 s followed by a latency period of another 45 s. In contrast, average milk flow reached its lowest values at a pre-stimulation of 15 s without additional latency period. However, average milk flow after a 15-s pre-stimulation increased with increasing latency period. In conclusion, a very short pre-stimulation when followed by a latency period up to 45 s before teat cup attachment remains a suitable alternative for continuous stimulation to induce milk ejection.

There is conflicting evidence in the literature as to whether up-regulation of urokinase plasminogen activator (u-PA) expression is related to bovine mammary epithelial cell growth. The role of u-PA receptor (u-PAR) and that of the plasminogen activator inhibitors type 1 and type 2 (PAI-1 and PAI-2) in bovine mammary epithelial cell proliferation is not known. The effect of growth factors and various hormones known to affect mammary function on expression of u-PA, u-PAR, PAI-1, PAI-2 and cell proliferation using the BME-UV1 bovine mammary epithelial cell line was examined. Cell proliferation was measured using the MTT assay and direct cell enumeration. Results showed that both IGF-1 and EGF increased cell proliferation but EGF was a more potent mitogen than IGF-1. Furthermore, IGF-1 increased by 2-fold expression of both u-PA and u-PAR while EGF increased by 3·8-fold the expression of only u-PAR. Both growth factors had no effect on expression of PAI-1 and PAI-2. In a manner consistent with changes in gene expression, EGF and to a lesser extent IGF-1 up-regulated total cell associated, membrane-bound and secreted u-PA activity. Thus, a strong correlation exists between u-PAR gene expression along with the activity of u-PA present on cell membranes and cell proliferation. Dexamethasone, prolactin and surprisingly insulin had no effect on cell proliferation. Dexamethasone alone and when combined with insulin or prolactin up-regulated gene expression of both PAI- and PAI-2 but not that of u-PA and u-PAR. Decreased total cell-associated, membrane-bound and secreted u-PA activity was detected in cells cultured in the presence of dexamethasone when combined with insulin or prolactin. However no such effect was observed in the presence of dexamethasone alone. Thus, dexamethasone acting synergistically with prolactin or insulin inhibits the activation of the plasmin-plasminogen system but this inhibition is not correlated with any changes in cell proliferation.

A method for evaluating cleaning effect based on Bacillus cereus spores was developed and tested in a model system designed to resemble actual farm conditions. A test rig with four removable sampling plates was mounted in a milk line. The plates were attached at the end of T-junctions protruding either 1·5 or 3-times the milk pipe diameter from the main loop to reflect different levels of cleaning difficulty. In each cleaning test, B. cereus spores were applied to the four sampling plates to simulate soil. A series of cleaning tests was conducted at 35, 45, 55 and 65°C with six commercial chlorine-free, alkaline detergents; three liquid and three powder-based products. A commercial alkaline detergent with chlorine, a sodium hydroxide solution, a sodium hydroxider/hypochlorite solution and pure water were also tested. Triplicate tests were performed with each cleaning solution, giving a total of 120 cleaning tests. The cleaning effect was evaluated by comparing the number of spores before and after cleaning. At all temperatures, the two chlorine-based cleaning solutions gave significantly greater reductions in B. cereus spores than the chlorine-free products. All six commercial chlorine-free, alkaline detergents generally gave similar cleaning effects, with no differences in the performance of powder-based and liquid forms. The mechanical spore reduction effect with water alone was greater (1·5–1·8 log-units) than the additional chemical effect of sodium hydroxide or chlorine-free detergents (0·5–1·2 log-units). The chlorine-based solutions had a considerably more powerful chemical effect (2–4 log-units depending on temperature). In general, an increase in cleaning solution temperature up to 55°C gave a greater reduction in spores. A further increase to 65°C did not improve cleaning effectiveness.

Aqueous solutions of reconstituted whey protein- concentrate (WPC) & isolate (WPI) powders were sonicated at 20 kHz in a batch process for 1–60 min. Sonication at 20 kHz increased the clarity of WPC solutions largely due to the reduction in the size of the suspended insoluble aggregates. The gel strength of these solutions when heated at 80°C for 20 min also increased with sonication, while gelation time and gel syneresis were reduced. These improvements in gel strength were observed across a range of initial pH values, suggesting that the mechanism for gel promotion is different from the well known effects of pH. Examining the microstructure of the whey protein gels indicated a compact network of densely packed whey protein aggregates arising from ultrasound treatment. Comparable changes were not observed with whey protein isolate solutions, which may reflect the absence of larger aggregates in the initial solution or differences in composition.

Selective dry cow therapy (SDCT) has received increasing attention in recent years owing to global concerns over agricultural use of antimicrobial drugs and development of antimicrobial resistance. The objective of this study was to evaluate the effect of SDCT on milk yield and somatic cell count (SCC) in dairy herds in the USA. Cows in four Ohio dairy herds were categorized into two groups (low-SCC and high-SCC) at dry-off based on their SCC and clinical mastitis (CM) history during the lactation preceding the dry-off. Low-SCC cows were randomly assigned to receive or not to receive intramammary antibiotics at dry-off. Milk yield and SCC of these cows during the following lactation were compared using linear mixed effects models, adjusting for parity, calving season, stage of lactation, previous lactation milk yield and herd. Milk yield of untreated and treated low-SCC cows at dry-off did not differ significantly during the following lactation. Overall, treated low-SCC cows had 16% lower SCC (approximately 35 000 cells/ml, P=0·0267) than the untreated cows during the following lactation; however, the effect was variable in different herds. Moreover the impact of treatment, or the lack thereof, on milk yield varied considerably between herds. The results suggested that in some herds treating all cows at dry-off may be beneficial while in other herds leaving healthy cows without antibiotic dry cow treatment has no negative impact on milk yield or milk quality (SCC), and in fact, may be beneficial. Further studies are needed to identify characteristics of herds where treating all cows routinely at dry-off may be needed for maintaining good udder health and where switching to selective treatment of cows at dry-off would be the optimal approach to achieve best results.