Ketone Bodies in Milk and Blood of Dairy Cows: Relationship between Concentrations and Utilization for Detection of Subclinical Ketosis

Transcription

1 J. Dairy Sci. 84: American Dairy Science Association, Ketone Bodies in and of Dairy Cows: Relationship between Concentrations and Utilization for Detection of Subclinical Ketosis F. Enjalbert,* M. C. Nicot,* C. Bayourthe, and R. Moncoulon *École Nationale Vétérinaire, Département Élevage & Produits, Laboratoire d Alimentation, 23 Chemin des Capelles, Toulouse Cedex 3, France École Nationale Supérieure Agronomique, Laboratoire d Ingénierie Agronomique, Avenue de l Agrobiopôle, BP 107, Auzeville-Tolosane, Castanet Tolosan Cedex, France ABSTRACT The objective of this study was to investigate the relationship between the concentrations of the different ketone bodies in milk and blood and to evaluate these concentrations for the detection of subclinical ketosis. A total of 60 multiparous cows were used. Concentrations of acetone, acetoacetate, and β-hydroxybutyrate were analyzed quantitatively in blood and milk, and the Ketolac strip test was used for semiquantitative determination of β-hydroxybutyrate in milk. Cows were defined subclinically ketotic when their concentration of blood β- hydroxybutyrate was over 1200 µmol/l. High correlation coefficients were observed between blood acetone and blood acetoacetate, and between blood and milk acetone. On the contrary, concentrations of milk and blood β- hydroxybutyrate were poorly correlated with the other concentrations of ketone bodies. The Ketolac strip test overestimated the concentrations of β-hydroxybutyrate in milk. For the detection of subclinical ketosis, the best sensitivity-specificity combination was obtained with the determination of acetoacetate in blood or milk, with threshold concentrations of 125 and 50 µmol/l, respectively. Determination of β-hydroxybutyrate in the milk via an enzymatic analysis or via the Ketolac strip test provided valuable results, with threshold concentrations of 70 and 100 µmol/l, respectively. The simplicity of use of the Ketolac strip test makes it a valuable way to investigate subclinical ketosis. (Key words: subclinical ketosis, detection, ketone bodies, Ketolac) Abbreviation key: Ac = acetone, AcAc = acetoacetate, BAc = blood acetone, BAcAc = blood acetoacetate, BBHBA = blood BHBA, MAc = milk acetone, MAcAc = milk acetoacetate, MBHBA = milk BHBA. Received July 28, Accepted October 24, Corresponding author: F. Enjalbert; f.enjalbert@envt.fr. INTRODUCTION Ketosis is a common disease in high producing dairy cows. It is caused by a negative energy balance, and typically occurs within 2 mo after calving. Even when clinical signs do not appear, ketosis can affect milk production (Gustafsson et al., 1993) and reproduction (Andersson et al., 1991; Andersson and Emanuelson, 1985) and be associated with an increased frequency of left displaced abomasum (Geishauser et al., 1997), or decreased nonspecific immunity (Sartorelli et al., 2000). Because of the economic consequences, the detection of cows with subclinical ketosis is important in order to treat individual cows or to improve the diet. Clinical and subclinical ketosis both result in increased concentrations of ketone bodies in tissues and milk of the cows. BHBA (BBHBA) concentration has often been used for this detection, and several authors used a cut-off point of 1200 µmol/l to discriminate between healthy cows and animals with subclinical ketosis (Duffield et al., 1997, 1998; Geishauser et al., 1998; Jorritsma et al., 1998). However, blood sampling is not easy for farmers, and determination of ketone bodies in milk can make sampling easier. Moreover, semiquantitative cow-side tests for determination of milk BHBA (MBHBA) recently became available (Geishauser et al., 1998) and can provide an immediate result. Evaluations of these cow-side tests have already been published (Geishauser et al., 1998; Jorritsma et al., 1998), but the literature does not provide extensive data for comparing concentrations of ketone bodies in milk and blood or for comparing epidemiological interest of the different ketone bodies. The objectives of this study were 1) to investigate the relationship between the concentrations of acetone (Ac), acetoacetate (AcAc), and BHBA in milk and blood, 2) to assess the reliability of the BHBA concentrations as measured by a semiquantitative test, the Ketolac strip test (Hoechst Roussel Vet, F Pantin Cedex, France), and 3) to propose threshold values for detection of cows with subclinical ketosis using Ac, AcAc, or BHBA concentrations in blood and milk, or MBHBA concentration as determined via the Ketolac test. 583

2 584 ENJALBERT ET AL. MATERIALS AND METHODS Animals A total of 60 Holstein cows were used in this study; 40 of them were from an experimental station, and 20 of them were in commercial dairy herds with a high prevalence of ketosis. All diets were based on corn silage. Primiparous cows were not used, and the frequency of distribution of lactation number of cows is represented on Figure 1. In the experimental station, all multiparous cows were sampled during the second and fourth weeks of lactation. Commercial dairy herds were visited only once for screening of cows, and all cows that were less than 6 wk after calving were sampled. Cows with BBHBA concentration over 1200 µmol/l were considered to suffer from subclinical ketosis, and denoted as true positive. In both kinds of herds, true positive cows were sampled again 3 d later, so that a total of 125 samples were collected, and individual cows contributed one (healthy cows in commercial dairy herds) to four (cows detected postitive during both second and fourth weeks of lactation in the experimental station) samples. Figure 2 shows the frequency of distribution of week of lactation of the donor cows of these 125 samples. Samples and Analyses On each cow, four coccygeal blood samples were taken before the afternoon milking into 10-ml glass tubes. Three of these tubes had previously been filled with 4- ml of perchloric acid 0.7 N, so that blood proteins precipitated during sampling (Andre and Coppin, 1982). These three tubes were strongly stirred after blood sampling, and weighed to determine the exact amount of blood. According to the result, perchloric acid, or blood from the fourth tube, was added to obtain exactly 50% blood Figure 2. Frequency of distribution of week of lactation of donor cows for the 125 samples. and 50% perchloric acid. Tubes were refrigerated at 4 C, and within 2 h were centrifuged at 3000 g for 4 min at 4 C. The supernatant was transferred into 5-ml tubes, and frozen at 30 C until analysis. About 100 ml of milk was sampled during the beginning of the afternoon milking. A semiquantitative determination of MBHBA concentration with the Ketolac BHB strip was performed immediately. Results were denoted as 0 to 49, 50 to 99, 100 to 199, 200 to 499, and 500 according to the MBHBA concentration (µmol/l). Subsequently, 15 ml of milk was deproteinized with 15 ml perchloric acid 0.7N, centrifuged, and stored as described above for blood samples. For Ac and AcAc determination, blood and milk samples were neutralized with a 3 M potassium phosphate buffer. Final ph was between 6.5 and 7.0. Samples were immediately placed into a bath of ice for 30 min to precipitate potassium perchlorate without transformation of AcAc into Ac and centrifuged (Andre and Coppin, 1982). Acetone was analyzed by GLC (Dani GC 1000, equipped with flame ionization detector and split injector; Viale Elvezia, 42-I Monza, Italy, column SGE BP20 1.0, SGE SARL, BP50, F Villeneuve Saint Georges Cedex, France), with N-propanol as an internal standard. Acetoacetate was determined by an enzymatic method (Boehringer Mannheim, Melan, France). An enzymatic kit ( ; Boehringer Mannheim) was used for BHBA determination. Epidemiological Study Figure 1. Frequency of distribution of lactation number in the 60 cows. For each ketone body in milk or blood, we tested a wide range of threshold concentrations, defining positive cows as cows with ketone body concentration over threshold. True positive cows being defined as cows with BBHBA concentration over 1200 µmol/l, for each ketone body and each threshold concentration, we computed

3 KETONE BODIES FOR DETECTION OF SUBCLINICAL KETOSIS 585 Figure 3. Frequency of distribution of the concentration of BHBA (µmol/l) in the 125 blood samples. sensitivity (proportion of true positive cows testing positive), specificity (proportion of true negative cows testing negative), positive predictive value (proportion of cows testing positive that were true positive), and negative predictive value (proportion of cows testing negative that were true negative). Assuming that the economic cost associated with false negative (1-sensitivity) is higher than the cost associated with false positive (1-specificity), we chose threshold values for each test to obtain a sensitivity over 90%. RESULTS AND DISCUSSION Figure 3 represents the frequency of distribution of blood samples according to BBHBA concentration. Of the 125 samples studied, 19.2% were from cows with subclinical ketosis, but because of the limited number of cows used in this study, this value cannot be considered as representative of the south of France. Because a number of the cows used were in herds with known high prevalence of ketosis, and because true positive cows were tested again 3 d later but true negative cows were not, this value probably overestimates the prevalence of subclinical ketosis. Using the same threshold, previous studies have reported 16.4% of positive cows in Ontario (Geishauser et al., 1998) and 14.0% in Netherlands (Jorritsma et al., 1998). Prevalence of subclinical ketosis increases from primiparous to multiparous cows (Detilleux et al., 1994), and is highest during lactation wk 2 and 3 (Duffield et al., 1998; Geishauser et al., 1998), so that the choice of cows submitted to tests can greatly influence the results. In the study in Ontario (Geishauser et al., 1998), cows of all parities were tested between lactation wk 1 and 9. In the study in Netherlands (Jorritsma et al., 1998), only cows of parity two or higher were tested, between 35 and 50 d after calving. Table 1 presents the mean concentrations of Ac, AcAc, and BHBA in blood and milk in true negative and true positive cows, and Table 2 presents the correlation coefficients observed between concentrations of ketone bodies in blood and milk. AcAc and blood Ac (BAc) were strongly correlated (r = 0.80), with a mean ratio of BAc to blood AcAc (BAcAc) of Andersson (1984) found a similar correlation coefficient between these two blood ketone bodies (r = 0.82), and a BAc/BAcAc ratio of The relationships between BBHBA and BAc or BAcAc were not as strong. The low correlation coefficient between BBHBA and BAc contrasts with the results of Andersson (1984), who found a strong relationship (r = 0.81). The correlation coefficient between BBHBA and BAcAc observed in the present study was in accordance with the 0.74 value published by Andersson (1984), and lower than the 0.87 value published by Kauppinen (1983). However, this last author sampled blood from the mammary vein, where the concentrations of ketone bodies are much lower than in jugular or coccygeal blood, because of mammary uptake. Moreover, mammary blood presents a different Table 1. Concentrations of ketone bodies (µmol/l) in blood and milk from true negative and true positive cows. True negative 1 True positive 2 (n = 101) (n = 24) All cows Mean SEM Mean SEM Mean SEM Acetone Acetoacetate BHBA Acetone Acetoacetate BHBA True negative cows: cows with BHBA concentration in blood lower than 1200 µmol/l. 2 True positive cows: cows with BHBA concentration in blood over 1200 µmol/l.

4 586 ENJALBERT ET AL. Table 2. Correlation coefficients between blood and milk concentrations of ketone bodies [n = 125, all values were significant (P < 0.001)]. Acetone Acetoacetate BHBA Acetone Acetoacetate BHBA Acetone 1.00 Acetoacetate BHBA Acetone Acetoacetate BHBA pattern of ketone bodies, because in the study of Kauppinen (1983), the mean BAcAc/BBHBA ratio was 0.38, whereas it was 0.15 in our work and 0.13 in the study of Andersson (1984). In our study, the mean BAcAc/BBHBA ratio was not different between healthy cows and cows with subclinical ketosis. This result agrees with the finding of Kauppinen (1983), who found no variation of the BAcAc/BBHBA ratio in mammary blood for BBHBA concentrations ranging from 200 to 5500 µmol/l. On the contrary, Filar (1979) found that the BAcAc/BBHBA ratio increased from about 0.08 in healthy cows (mean BBHBA 430 µmol/l) to 0.14 in cows with subclinical ketosis (mean BBHBA 1250 µmol/l), and 0.24 in ketotic cows (mean BBHBA 3270 µmol/l). A similar trend was found by Andersson (1984). Cows in negative energy balance oxidize large amounts of fatty acids in the liver cells, which leads to a high NADH/NAD ratio in the mitochondria. However, in the liver of ruminants, conversion of BHBA to AcAc, which needs NAD, does not depend on the mitochondrial NADH/NAD ratio because this conversion is cytosolic (Koundakjian and Snoswell, 1970). In ketotic cows, the cytosolic NADH/NAD ratio is low (Heitmann et al., 1987), so that BHBA is converted to AcAc, which explains that in ketosis-induced cows, the AcAc/BHBA ratio in the liver cells increased (Mills et al., 1986). However, this BAcAc/BBHBA ratio also depends on the body condition of the cows, because after induction of ketosis, obese cows had a lower BAcAc/BBHBA ratio than normal cows (Smith et al., 1997). The mammary metabolism could also be involved in the equilibrium between BAcAc and BBHBA, via a net release of AcAc resulting from conversion of BHBA (Bergman, 1971). Such an activity of the mammary gland could depend on the milk production, and explain the discrepancies between experiments. The correlation coefficient between the three milk ketone bodies ranged from 0.68 to Much lower values, varying from 0.11 to 0.43, were published by Andersson (1984), but with a similar ratio of milk Ac (MAc) to milk AcAc (MAcAc). Ac was very close to BAc, with a mean MAc/BAc ratio of Similarly, Andersson (1984) observed such a close relationship (r = 0.96), but with a mean MAc/BAc ratio of The relationship between concentrations in blood and milk was not as strong for AcAc and BHBA. Concentrations in milk were much lower than concentrations in blood, and BAcAc and BBHBA explained only 55 and 43% of the variations of MAcAc and MBHBA, respectively. Andersson (1984) also found a low correlation coefficient between BAcAc and MAcAc (r = 0.45) and no relationship between MBHBA and BBHBA (r = 0.01). In a recent paper, Geishauser et al. (1998) reported from literature that correlation coefficients between BBHBA and MBHBA range, according to author, from 0.00 to Because BHBA can be used by the mammary gland for fatty acids synthesis and AcAc can be converted to butyrate (Dodds et al., 1981), lower concentrations in milk than in blood could be expected. Our Table 3. BHBA concentrations measured for the 5 different values of Ketolac strip test. BHBA concentration, µmol/l Ketolac class (µmol/l) n Mean SEM Minimum Maximum

5 KETONE BODIES FOR DETECTION OF SUBCLINICAL KETOSIS 587 mean MAcAc/BAcAc and MBHBA/BBHBA ratios (0.40 and 0.12, respectively) were similar to values (0.45 and 0.13, respectively) reported by Andersson (1984). Mean MBHBA concentrations for the different values of the Ketolac test are reported on Table 3. Each class of Ketolac test corresponded with a very wide range of values as measured by enzymatic test, and the Ketolac test tended to overestimate the true MBHBA concentration. For Ketolac class 100 to 199, the mean true concentration of MBHBA was 102 µmol/l, which is the lower limit of the range of Ketolac values. The overestimation was most important for the Ketolac class 500 µmol/l. In previous evaluations of the Ketolac test, no quantitative determination of MBHBA was performed (Geishauser Figure 4. Sensitivity ( ), specificity ( ), positive predictive value (+), and negative predictive value ( ) of blood and ketone bodies for the detection of subclinical ketosis (>1200 µmol of BHBA/L of blood).

6 588 ENJALBERT ET AL. Table 4. Best threshold, sensitivity, specificity, positive predictive value (pv+), and negative predictive value (pv ) of blood and milk ketone bodies, assuming that true positive were animals with blood BHBA concentration >1200 µmol/l. Best threshold 1 Sensitivity Specificity pv+ pv µmol/l % % % % Acetone Acetoacetate Acetone Acetoacetate BHBA Ketolac Best threshold: limit concentration to decide that a cow is positive, with a sensitivity over 90%. et al. 1998, 2000; Jorritsma et al, 1998) for comparison of values. Sensitivity, specificity, positive predictive value, and negative predictive value of BAc, BAcAc, MAc, MAcAc, MBHBA, and Ketolac test, with BBHBA as a gold standard at a cut-off point of 1200 µmol/l, are presented in Figure 4. Increasing threshold concentration of ketone bodies resulted in a decreased sensitivity and an increased specificity. The intersection between the curves of sensitivity and specificity was high for BAcAc, which is indicative of a good epidemiological value. Moreover, use of BAcAc provided high positive predictive values. To our knowledge, such values have not been published for BAc, BAcAc, MAc, MAcAc, and MBHBA. Geishauser et al. (1998) and Jorritsma et al. (1998), with the same gold standard, found for 100 to 199 µmol Ketolac BHBA/ L lower sensitivities (72.4 and about 80%, respectively) and higher specificities (89.4 and about 85%, respectively). These differences could be due to analytical method: the method for quantitative determination of BBHBA was not indicated in the study of Jorritsma et al. (1998), and the exact method for determination of BBHBA was not clear in the study of Geishauser et al. (1998) because the β-hba 310-UV kit from Sigma is designed to be used with serum or plasma, not with whole blood. Table 4 presents the best threshold values for detection of ketosis via determination of BAc, BAcAc, MAc, MAcAc, MBHBA, and Ketolac test. Values for quantitative tests were lower than values resulting from the ratios between concentrations in Table 1 and a BBHBA concentration of 1200 µmol/l, because we gave some priority to sensitivity over specificity, and sensitivity decreases when the threshold value of a test increases. For all criteria, negative predictive values were over 96%. The different tests gave different values for specificity and positive predictive value. The determination of BAcAc and MAcAc gave the best specificities and positive predictive value. Kauppinen (1983) already stated that BAcAc was a good test for diagnosis of ketosis, because its concentration shows less diurnal variation than BBHBA values. Our proposed threshold of 125 µmol/l for BAcAc is consistent with previously observed concentrations of 30 ± 10 µmol/l in healthy cows, and 180 ± 10 µmol/l in cows with subclinical ketosis (Filar, 1979). The corresponding threshold of 50 µmol/l for MAcAc is compatible with the BAcAc threshold and the mean MAcAc/BAcAc ratio observed in the present study. The limit in routine use of BAcAc or MAcAc is the great instability of this compound, which requires rapid precipitation of proteins and freezing of samples (Työppönen and Kauppinen, 1980). The use of BAc for testing ketosis resulted in higher specificity than the use of MAc, with a slightly higher threshold. Very different thresholds have already been proposed for MAc. Miettinen (1994), based on a study of the relationship between MAc and milk production proposed an upper limit of 50 µmol/l. A threshold of 400 µmol/l was used in a study on fertility (Andersson and Emanuelson, 1985) and in a study on milk production (Gustafsson et al., 1993). Steen et al. (1996) classified cows using 700 µmol/l MAc as a limit. Later, studying the effects of ketonemia on milk production on a large set of field data, Gustafsson and Emanuelson (1996) suggested that 1400 µmol/l should be used as a critical value. Because the MAc/BBHBA ratio was about 0.27 in our study, such a threshold can be seen as very high. Moreover, Filar (1979) reported BAc concentrations of 70 ± 20 µmol/l in healthy cows, and 420 ± 70 µmol/l in cows with subclinical ketosis, which bracket our limit of 175 µmol/l. The use of MBHBA to discriminate ketotic cows proved to be nearly as reliable as BAc. Lack of relationship between MBHBA and BBHBA observed by Andersson (1984) suggested that MBHBA could be of limited value for the detection of subclinical ketosis, so that few authors have presented a critical concentration for MBHBA. Geishauser et al. (98) indicated that 100 µmol/

7 KETONE BODIES FOR DETECTION OF SUBCLINICAL KETOSIS 589 L could be the limit between normal and ketotic cows, which is higher than our proposed value (70 µmol/l). Determination of MBHBA concentration via the semiquantitative Ketolac strip test provided similar sensitivity, specificity, and predictive values than enzymatic chemical analysis, when used with a critical value of 100 to 199 µmol/l, which is coherent with the overestimation of MBHBA when measured via the test. Jorritsma et al. (1998) also assessed that 100 to 200 MBHBA µmol/l measured via the Ketolac test resulted in the most desirable sensibility-specificity combination. Geishauser et al. (1998) proposed a cut-off point of 50 or 100 µmol/l for an individual purpose, and 200 or 500 µmol/l for making changes in the diet, and after a further study (Geishauser et al., 2000), proposed 200 µmol/l as a reasonable minimal concentration for detection of cows with subclinical ketosis. CONCLUSIONS Concentrations of milk and blood ketone bodies were fairly well related, and semiquantitative determination via the Ketolac strip test provided valuable results. Because subclinical ketosis can be due to individual factors, i.e., level of intake or level of production, and to herd factors, i.e., diet during the dry period and the beginning of lactation, individual detection of ketosis can be performed via quantitative or semiquantitative determination of blood or milk ketone bodies. Positive cows have to be treated, and a correction of the diet is necessary when a high prevalence of positive cows is encountered. ACKNOWLEDGMENT The authors acknowledge Hoechst Roussel Vet France for financial support and supply of Ketolac strip tests. REFERENCES Andersson, L Concentrations of blood and milk ketone bodies, blood isopropanol and plasma glucose in dairy cows in relation to the degree of hyperketonaemia and clinical signs. Zentralbl. Veterinaermed. Reihe A 31: Andersson, L., and U. Emanuelson An epidemiological study of hyperketonaemia in swedish dairy cows; determinants and the relation to fertility. Prev. Vet. Med. 3: Andersson, L., A. H. Gustafsson, and U. Emanuelson Effect of hyperketonaemia and feeding on fertility in dairy cows. Theriogenology 36: André, F., and Coppin, F Etude technique du dosage de l hydroxybutyrate et de l acetoacetate dans le sang et le lait des bovins. Ann. Rech. Vet. 13: Bergman, E. N Hyperketonemia-ketogenesis and ketone body metabolism. J. Dairy Sci. 54: Detilleux, J. C., Y. T. Gröhn, and R. L. Quaas Effects of clinical ketosis on test milk yields in finnish Ayrshire cattle. J. Dairy Sci. 77: Dodds, P. F., M. G. Guzman, S. C. Chalberg, G. J. Anderson, and S. Kumar Acetoacetyl-CoA reductase activity of lactating bovine mammary fatty acid synthase. J. Biol. Chem. 256: Duffield, T. F., D. F. Kelton, K. E. Leslie, K. D. Lissemore, and J. H. Lumsden Use of test day milk fat and milk protein to detect subclinical ketosis in dairy cattle in Ontario. Can. Vet. J. 38: Duffield, T. F., D. Sandals, K. E. Leslie, K. Lissemore, B. W. McBride, J. H. Lumsden, P. Dick, and R. Bagg Efficacy of monensin for the prevention of subclinical ketosis in lactating dairy cows. J. Dairy Sci. 81: Filar, J Über den Gehalt an β-hydroxybutyrat, Azetazetat and Azeton im Blut von gesunden und an Ketose erkrankten Kühen. Wien. Tierärztl. Monatschr. 66: Geishauser, T., K. Leslie, T. Duffield, and V. Edge An evaluation of milk ketone tests for the prediction of left displaced abomasum in dairy cows. J. Dairy Sci. 80: Geishauser, T., K. Leslie, D. Kelton, and T. Duffield Evaluation of five cowside tests for use with milk to detect subclinical ketosis in dairy cows. J. Dairy Sci. 81: Geishauser, T., K. Leslie, J. Tenhang, and A. Bashiri Evaluation of eight cow-side ketone tests in milk for detection of subclinical ketosis in dairy cows. J. Dairy Sci. 83: Gustafsson, A. H., L. Andersson, and U. Emanuelson Effect of hyperketonaemia, feeding frequency and intake of concentrate and energy on milk yield in dairy cows. Anim. Prod. 56: Gustafsson, A. H., and U. Emanuelson acetone concentration as an indicator of hyperketonaemia in dairy cows: the critical value revised. Anim. Sci. 63: Heitmann, R. N., D. J. Dawes, and S. C. Sensenig Hepatic ketogenesis and peripheral ketone body utilization in the ruminant. J. Nutr. 117: Jorritsma, R., S. J. C. Baldée, Y. H. Schukken, Th. Wensing, and G. H. Wentink Evaluation of a milk test for detection of subclincal ketosis. Vet. Quart. 20: Kauppinen, K Correlation of whole blood concentrations of acetoacetate, β-hydroxybutyrate, glucose and milk yield in dairy cows as studied under field conditions. Acta Vet. Scand. 24: Koundakjian, P. P., and A. M. Snoswell Ketone body and fatty acid metabolism in sheep tissues: 3-hydroxybutyrate dehydrogenase, a cytoplasmic enzyme in sheep liver and kidney. Biochem. J. 119: Miettinen, P.V.A Relationship between milk acetone and milk yield in individual cows. J. Vet. Med. Ser. A 41: Mills, S. E., D. C. Beitz, and J. W. Young Characterization of metabolic changes during a protocol for inducing lactation ketosis in dairy cows. J. Dairy Sci. 69: Sartorelli, P., S. Paltrinieri, and S. Comazzi Non-specificimmunity and ketone bodies. II; in vitro studies on adherence and superoxide anion production in ovine neutrophils. J. Vet. Med. Ser. A 47:1 15. Smith, T. R., A. R. Hippe, D. C. Beitz, and J. W. Young Metabolic characteristics of induced ketosis in normal and obese dairy cows. J. Dairy Sci. 80: Steen, A., O. Østeras, and H. Grønstøl Evaluation of additional acetone and urea analyses, and of the fat-lactose-quotient in cow milk samples in the herd recording system in Norway. J. Vet. Med. Ser. A 43: Työppönen, J., and K. Kauppinen The stability and automatic determination of ketone bodies in blood samples taken in field conditions. Acta Vet. Scand. 21:55 61.