Available online at www.sciencedirect.com

The Veterinary Journal The Veterinary Journal 176 (2008) 10–20 www.elsevier.com/locate/tvjl

Nutritional and management strategies for the prevention of fatty liver in dairy cattle Ric R. Grummer * Department of Dairy Science, University of Wisconsin, Madison, 1675 Observatory Dr. Madison, WI 53706, USA Accepted 18 December 2007

Abstract Fatty liver occurs in dairy cattle during periods of elevated blood non-esterified fatty acids (NEFAs). Elevated blood NEFAs are associated with hormonal changes at parturition and negative energy balance. Approaches for preventing fatty liver include inhibition of fatty acid mobilization from adipose tissues and altering hepatic metabolism to enhance fatty acid oxidation or export as a constituent of very low-density lipoproteins (VLDL). Nutritional and management strategies to implement these approaches have been examined. Increasing energy density of diet, either by increasing non-fiber carbohydrate or fat, has failed to prevent fatty liver. Two nutritional supplements, ruminally-protected choline and propylene glycol, have proven effective at preventing fatty liver. Choline probably enhances hepatic VLDL secretion. Propylene glycol most likely reduces fatty acid mobilization from adipose tissue. Shortening or eliminating the dry period is a management strategy that reduces the magnitude of negative energy balance after calving and triglyceride accumulation in the liver. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Fatty liver; Non-esterified fatty acids; Choline; Propylene glycol; Dry period length

Introduction A basic understanding of the etiology of fatty liver is imperative to appreciate a discussion of the strategies to prevent it. The etiology has been reviewed elsewhere (Grummer, 1993; Drackley et al., unpublished data). In brief, fatty liver occurs during times of elevated non-esterified fatty acid (NEFA) concentrations in the blood. The elevation may be a response to negative energy balance or to hormonal changes that accompany parturition. Hepatic uptake of NEFAs is related to blood flow to the liver and concentration of NEFAs in blood. If NEFA uptake exceeds the capacity of the liver to completely oxidize NEFAs to CO2, partial oxidation to form ketones or esterification to form triglycerides (TGs) may result. Production of ketones in moderation is acceptable because they can serve as an energy source for many tissues but *

Tel.: +1 608 263 3492; fax: +1 608 263 9412. E-mail address: [email protected].

1090-0233/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.tvjl.2007.12.033

excessive production can adversely affect animal behavior and performance. Esterification to form TGs is acceptable if they are exported as very low-density lipoproteins (VLDLs). Unfortunately, ruminants have a very slow rate of VLDL export from the liver, so the capacity for that route of NEFA disposal is easily exceeded. Approaches to prevent or treat fatty liver can be subdivided into three main categories: (1) reduce blood NEFAs by decreasing TG lipolysis in adipose tissue; (2) increase complete hepatic oxidation of NEFAs, or (3) increase the rate of VLDL export from the liver. The drawback of the first strategy is that it impedes a process that is intended to support lactation. Effectively, it shifts a ‘bottleneck’ from the liver to adipose tissue in an attempt to maintain a healthy, functional liver. The second strategy probably has limitations as well. Complete oxidation yields ATP, the energy currency of cells, of which there is a finite requirement. If complete oxidation is to continue beyond that necessary to provide energy for maintenance of the liver, it must be uncoupled from ATP production, which

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20

Nutritional strategies Nutritional practices to prevent or treat fatty liver can be divided into two main categories: (1) diet formulation to increase energy density and (2) inclusion of feed additives to modify metabolism in a manner to reduce the likelihood of liver TG accumulation. Diet formulation to increase energy density is typically done to minimize the magnitude of negative energy balance and reduce fatty acid mobilization from adipose tissue. Because energy balance of pre-fresh transition cows and early lactation dairy cows is often misunderstood, it will be reviewed prior to discussing the formulation of diets to increase energy density. Energy balance of pre-fresh transition and early lactation dairy cows Energy balance is the difference between energy consumed and the energy required for maintenance, growth, pregnancy, and lactation. The growth requirement for mammary tissue is not included in the most recent NRC (2001) for dairy cattle. This does not imply that it is trivial; it simply reflects the lack of data for determining the requirement. It is often stated that the energy status of pre-fresh transition cows is compromised because energy intake is decreasing due to declining dry matter intake (DMI), and energy requirements are increasing due to the growing fetus. This is probably not normally the case. Fig. 1 shows an estimate for the energy requirements for maintenance and growth of heifers and mature cows (NRC, 2001) and the change in energy requirements for the conceptus (NRC, 2001) and mammary growth (Grummer, unpublished data) during the last 60 days of gestation, i.e., the dry period. While the energy requirements for the conceptus and mammary growth are not trivial, the change in requirements during the final 60 days of gestation are quite small. Decreases in DMI prior to calving are typically 30– 35% (Hayirli et al., 2002) and are most dramatic during the final week prior to calving. Therefore, if energy status is to

14

11.6 12

Mcal NEl/day

9.7 10 8 6 4

3.4

3.7

2.9

-60d

-21d

-1d

2 0 Heifer

Cow

Conceptus + mammary growth during gestation

Maintenance + growth

Fig. 1. Average net energy (NEl) requirements per day for maintenance (and growth for heifers) versus growth of the conceptus and mammary tissue at various time points during the dry period. All estimates were from NRC (2001) except mammary growth, which was estimated (Grummer, unpublished data).

be compromised pre-calving, it will be due to changes in feed intake rather than changes in energy requirements. Energy intake, requirements, and balance of a group of cows that recently completed one of our experiments are shown in Fig. 2. It is important to note that even with the decline in energy intake associated with depressed feed intake prior to calving, the cows never experienced negative energy balance before they calved. Cows on this study were fed a diet containing 1.50 Mcal/kg DM during the far-off dry period and were fed a transition diet containing 1.69 Mcal/kg DM during the final 4 weeks prior to calving. In this study, dry cows were fed in excess of their energy requirements (NRC, 2001). Energy requirements of dry cows are extremely low, and except for the final few days prior to calving, can be met by feeding diets containing low quality forages and no concentrate. The data in Fig. 2 are an average for a group, which does not rule out that individual cows may experience greater degrees

Net energy for lactation (Mcal/day)

results in energy being lost as heat. The third strategy of increasing VLDL export is the most logical but little is know about what limits VLDL export in ruminants. This article will review the nutritional and management practices that our research laboratory and others have examined for employing these approaches for prevention and treatment of fatty liver. Particular attention is paid to studies in which lipid or TG accumulation in the liver was measured. This review will focus on the pre-fresh transition period (3–4 weeks pre-partum) through early lactation. Nutritional strategies focusing on the far-off dry period or entire dry period will be covered elsewhere (Drackley et al., unpublished data). Hormonal or pharmaceutical strategies for prevention and treatment of fatty liver will not be discussed.

11

50.0 40.0 30.0 20.0 10.0 0.0 -10.0-56

-49

-42

-35

-28

-21

-14

-7

0

7

14

21

-20.0

Day relative to calving Fig. 2. Energy (Mcal NEl/day) required (–) and consumed (4), and energy balance (d) for cows during an experiment. Cows were fed a diet containing 1.50 Mcal/kg DM during the far-off dry period and a ‘transition’ diet containing 1.69 Mcal/kg DM during the final 4 weeks prior to calving. After calving, the lactation diet contained 1.71 Mcal/kg (the small increase relative to transition diet is due to greater DMI following calving).

12

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20

of feed intake depression and negative energy balance prior to calving. The DMI of pre-fresh cows is extremely variable (Hayirli et al., 2002). As depicted in Fig. 2, early lactation is the period of greatest negative energy. Ironically, this is not the period during which blood NEFAs are highest. Non-esterified fatty acid concentrations peak at calving (Vazquez-Anon et al., 1994), which confirms that endocrine changes at calving along with negative energy balance contribute to adipose tissue lipolysis. A literature survey of 20 studies involving 52 dietary treatments indicated that the mean duration of negative energy balance post-partum is 45 days with a standard deviation of 21 days (Grummer and Rastani, 2003). For 90% of the treatments, cows reached energy balance by 63 days post-partum. Although there was a correlation between daily milk yield and energy balance (r = 0.26, P < 0.0001), the correlation with energy intake was considerably higher (r = 0.58, P < 0.0001). Increasing nutrient density of transition diets Boutflour (1928) at the World Dairy Congress first proposed the ‘steam up’ ration as a way to circumvent ‘the neglect of the preparation of the cows for her lactation period’. The term was meant to be an analogy to the preparation of a steam thresher. The concept of increasing grain feeding to transition dairy cows is still employed by many dairy producers. Initially, additional grain feeding was promoted as a means to acclimate the rumen microorganisms to additional non-fiber carbohydrate (NFC) that would be fed post-partum. Presumably, this would reduce the likelihood of acidosis and going off feed. Remarkably, this concept has never been experimentally tested so the validity is in doubt. If concentrates are fed separately from forages,

the need for microbial adaptation may be greater than when they are fed as a totally mixed ration. Feeding a totally mixed ration would allow for a more continuous consumption of modest amounts of grain, thereby minimizing wide fluctuations in pH and possibly in microbial populations. Also, the gradual increase in feed intake post-calving allows time for adaptation to increased concentrate feeding. More recently, additional grain feeding during the dry period has been promoted as a means to enhance papillae growth in the rumen (Dirksen et al., 1985). This would lead to more surface area for absorption of volatile fatty acids and minimize the likelihood of ruminal acidosis post-partum when grain feeding increases dramatically. Several recent studies have suggested that this is unlikely under normal dry cow and fresh cow feeding schemes (Ingvartsen et al., 2001; Friggens et al., 2004; Reynolds et al., 2003). Feeding additional grain prior to calving has been promoted as a means to reduce lipid-related metabolic disorders such as fatty liver (Grummer, 1993). This could occur by a couple mechanisms. Firstly, feeding additional grain leads to greater propionate production in the rumen. Propionate is an insulin secretagogue; insulin is antilipolytic and has the potential to decrease adipose tissue lipolysis. Additionally, increasing grain in the diet could increase its digestibility and, therefore, increase DM and energy intake. Table 1 summarizes nine studies that have been published in the last 10 years in which NFC has been increased in dry cow diets. Across trials, a wide variation in dietary NFC was tested. In eight of the studies, pre-partum DMI was measured and in five of those, DMI was significantly increased. Post-partum DMI was measured in seven of the studies, and milk yield was measured in eight of the

Table 1 A summary of recent trials examining the feeding low or high NFC diets beginning at dry-off (Grum et al., 1996; Douglas et al., 2004) or 3–4 weeks prior to parturition (all other studies listed below) Trial

NFC,% DM

NDF,% DM

Pre-partum DMI, kg/day

Post-partum DMI, kg/day

Milk yield, kg/day

Liver TG, units vary

Grum et al. (1996)

18 28 35 44 35 38 13 28 25 30 24 30 38 45 24 31 34 40

60 50 49 30 39 35 61 47 44 39 52 44 40 32 47 41 44 37

9.7 11.6* 10.2 13.0*

17.9 18.7

35.1 35.5

5.9 7.3 6.2 5.6

15.2 15.1 21.2 20.5 13.3 14.4 16.2 16.7 21.1 20.9 19.2 18.7

37.4 37.4 27.4 28 35.8 29.9 30.9 33.8 41.4 39.4 40.2 40.0 41.8 40.6

Minor et al. (1998) Mashek and Beede (2000) Keady et al. (2001) Holcomb et al. (2001) Doepel et al. (2002) Rabelo et al. (2003, 2005) Douglas et al. (2004) Smith et al. (2005)

9.28 11.03* 10.7 14.1* 13.9 12.8 11.3 13.0* 15.1 13.9 13.8 13.7

Values with a * indicate a significant difference (P < 0.05) between low and high NFC diets.

9.8 7.1* 9.2 8.7 5.4 7.6

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20

studies. There were no significant effects of NFC level on these two parameters in any of the studies. In only 1/5 studies in which liver TGs were measured post-partum was there a significant effect of increasing pre-partum NFC on decreasing liver TGs. Why is feeding high NFC diets not more beneficial in decreasing liver TGs? Friggens et al. (2004) suggested that lipid mobilization is a natural phenomenon common to all mammals and is part of an orchestrated pattern of bodyweight change to support lactation. They further suggested that feeding additional grain would not reduce the ‘normal’ amount of lipid that is mobilized. If the cow ‘normally’ does not experience appreciable negative energy balance prior to calving, then it seems reasonable to believe that there would be negligible effects on lipid mobilization and development of fatty liver prior to calving. If milk yield or post-partum feed intake is not affected by pre-partum dietary NFC, there should be no effect on energy balance and lipid mobilization post-partum. The only repeatable effect of increasing pre-partum diet NFC is increasing pre-partum DMI. However, recent summaries of data (Grummer et al., 2004) indicate that prepartum DMI has little relationship to post-partum liver TGs. The magnitude of decrease in pre-partum DMI is more closely related to plasma NEFAs and liver TGs in the first days following calving. That suggests that the change in energy balance prior to calving is a more important signal for lipolysis than energy balance per se. Since feeding additional NFC prior to calving increases DMI, we have speculated that it could possibly predispose cows to development of fatty liver because there is greater potential for a more significant drop in feed intake prior to calving. The data in Table 1 do not support that hypothesis. Holtenius et al. (2003) suggested that feeding high grain diets pre-partum may lead to prolonged elevation in blood insulin. This in turn could lead to insulin resistance, which would act to elevate lipolysis and increase the likelihood of fatty liver. This theory seems reasonable; unfortunately, the effect of diet on development of insulin resistance in pre-partum dairy cows has not been studied. An important question is how long would one need to feed extra grain for the cow to develop insulin resistance? There are data which suggest that feeding excess grain during the dry period for 6 weeks or longer may be deleterious to the cow and cause TG accumulation in the liver (e.g., Rukkwamsuk et al., 1999; Drackley et al., unpublished data). Increasing grain feeding for shorter periods of time, such as for a 3–4 week pre-fresh transition period, has not led to the development of fatty liver (Minor et al., 1998; Doepel et al., 2002; Rabelo et al., 2005). Unfortunately, there is very little information regarding the effects of increasing NFC of post-fresh transition diets on liver TGs. During the first 3 weeks after calving, there is extreme animal variation in metabolic parameters and, consequently, large animal numbers are needed to detect significant treatment effects on those parameters. Researchers are hesitant to design experiments that apply treatments

13

during that period of lactation. Rabelo et al. (2003, 2005) did employ a factorial design to examine the relative importance of pre- and post-fresh transition diets on accumulation of TGs in the liver. For the final 4 weeks prior to calving, cows were fed diets that contained 1.58 or 1.70 Mcal of NEl/kg DM. After calving, one half of the cows from each of these treatments received diets that contained 1.57 or 1.63 Mcal NEl/kg DM for 21 days. After that, all cows received a diet that contained 1.63 Mcal NEl/kg DM. (Diet NEl values were determined according to NRC (2001); therefore, the lower values post-partum reflect the effect of higher feed intake and lower diet digestibility and does not mean that they contained more fiber and less NFC than pre-partum diets.) There were no prefresh diet effects on liver TGs at days 1, 21, or 35 post-partum. The high energy diet post-partum did reduce liver TGs at day 21 post-partum in multiparous cows, however, that effect was no longer present at day 35 post-partum after all cows had received the same diet for 2 weeks. These data suggest that dietary NFC level immediately post-partum may have a greater influence on liver TGs than dietary level immediately pre-partum. Further studies are required to verify this observation. Supplementing dietary fat to increase dietary energy density Fat supplementation is another strategy for increasing energy density of transition diets. Kronfeld (1982) speculated that feeding supplemental fat would reduce fatty acid mobilization from adipose tissue. This strategy assumes dietary fatty acids are incorporated into intestinally synthesized lipoproteins that are metabolized predominantly by tissues other than the liver. In contrast, fatty acids mobilized from adipose tissue are extensively utilized by the liver. However, plasma NEFA concentrations almost always increase when supplemental fat is fed (Grummer and Carroll, 1991; Chilliard, 1993). This is probably the result of fatty acids ‘spilling’ into blood as tissues metabolize the lipoprotein. Also, several studies have indicated that supplemental fat actually increases adipose tissue lipolysis rather than decreases it (Chilliard, 1993; Kulick et al., 2006). Feeding supplemental fat to cows from 17 days pre-partum through 15 weeks post-partum did not reduce liver TGs or plasma beta-hydroxybutyrate during the transition period (Skaar et al., 1989). In fact, liver TGs tended to increase due to fat feeding. We conducted a follow-up study to examine the effects of fat on liver TGs during feed restriction of far-off dry cows (Bertics and Grummer, 1999). Feed was restricted (25% of ad libitum intake) for 10 days to stimulate fatty acid mobilization and mimic pre-partum feed intake depression. We used this model because the cows are less variable than transition cows and they can be utilized in an experimental design that allows each animal to serve as its own control. This increases our ability to detect treatment differences. Additionally, the model allows us to specifically examine devel-

14

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20

opment of fatty liver due to negative energy balance and eliminates the confounding effects of hormonal changes that accompany parturition and lactogenesis. Results indicated that fat supplementation increased NEFAs and liver TGs. We also investigated if fat feeding would influence the rate of TG depletion from the liver in cows that had fatty liver induced prior to resumption of ad libitum feeding and treatment (Bertics and Grummer, 1999). Feeding fat reduced the rate of TG depletion. Feeding fat during the transition period may acclimate cows to dietary fat and, therefore, reduce the likelihood of DMI depression that might occur if introduced post-partum. However, Salfer and coworkers (1995) did not observe any improvement in lactation performance of cows that were introduced to dietary fat pre-partum versus cows that received fat starting at parturition or 35 day post-partum. Allen et al. (1995) utilized a 2  2 factorial design to examine the effects of fat supplementation during pre-partum, post-partum, or pre- and postpartum on health, production and reproduction. Pre-partum diets were isocaloric with 75% forage and 25% concentrate. Pre-partum treatments were corn-soy versus wheat middling fat-based concentrates (1.5% of total DM as fat). After calving, one half of the animals from each pre-partum treatment were fed 0% or 4% animal fat. Feeding fat pre-partum did not affect DMI but did increase plasma NEFAs. Feeding fat after calving increased plasma NEFAs and beta-hydroxybutyrate concentrations, decreased plasma glucose, DMI, and milk yield, and increased the incidence of ketosis. There were no pre-partum by post-partum treatment interactions, which indicated that fat supplementation during the transition period did not help acclimate cows to fat supplementation post-partum. Specific fatty acids may be able to alter hepatic lipid metabolism such that TG accumulation is minimized (Mashek and Grummer, 2003). Because the mode of action is independent of an effect through reformulation of the diet for increased energy density and improved energy balance, it will be covered in the feed additive section below. Feed additives for prevention or treatment of fatty liver Feed additives can be classified into different categories depending on their intended mode of action: reduce adipose lipolysis, enhance hepatic VLDL secretion, or increase hepatic fatty acid oxidation. Reduction of adipose lipolysis Compounds that may decrease adipose tissue lipolysis include propylene glycol (PG), monensin, chromium (Cr), niacin, and conjugated linoleic acid (CLA). Positive effects of monensin in lowering blood ketones are well documented (Duffield et al., 1998; Zahra et al., 2006), but to my knowledge, only one study has examined its effects on liver TGs of transition dairy cows (Zahra et al., 2006). A

controlled-release capsule delivering 335 mg monensin/ day for 95 days was initiated at 3 weeks prior to calving; no effects on liver TGs were observed at 1 day or 3 weeks after calving. Although monensin lowered plasma betahydroxybutyrate concentration, there were no effects on plasma NEFA. An excellent and extensive review of PG effects on physiological responses of cows was recently published (Nielsen and Ingvartsen, 2004). Administering PG has the potential to cause an insulin response and reduce fatty acid mobilization from adipose tissue. PG drenched orally once daily (1 L) starting 10 days pre-partum until calving increased plasma glucose and insulin pre-partum and reduced total liver lipids and plasma NEFA immediately post-partum (Studer et al., 1993). Similarly, drenching 400 mL PG once daily from 7 days prior to expected calving until 7 days after calving reduced liver TG at 2 and 4 weeks post-calving (Rukkwamsuk et al., 2005). Because of the difficulty of drenching cows, there is interest in adding PG directly to the diet. Christensen et al. (1997) administered 300 mL PG once daily either as an oral drench, in 3 kg of concentrate, or in a total mixed ration for 7 days in feed-restricted springing heifers and cows. Liver TG was not measured; however, administering PG mixed with concentrate was nearly as effective as the oral drench for lowering plasma NEFA concentrations. Administering PG in a total mixed ration was not effective. This was probably the result of more continuous administration of low quantities of PG. Perhaps there never was sufficient PG administered over a narrow enough time frame to cause an insulin response. Consequently, oral drenching of PG appears more effective than feeding. Niacin has antilipolytic properties and has been investigated as a feed additive to prevent fatty liver and ketosis. Early studies indicated that niacin feeding could reduce blood ketones in ketotic cows (Waterman et al., 1972). However, there were no cows assigned to a control treatment (no niacin), and therefore niacin effects were confounded with time. Nevertheless, niacin became a major feed additive in dairy diets. A summary of the literature published in the most recent NRC (2001) for dairy cattle indicated little benefit of supplemental niacin (3–12 g/ day) in reducing plasma NEFAs or beta-hydroxybutyrate. Either pre- or pre- and post-partum supplementation of 6– 12 g niacin/day did not reduce liver TGs in transition dairy cattle (Skaar et al., 1989; Minor et al., 1998). In retrospect, that is not surprising because as much as 95% of dietary niacin is degraded in the rumen (Santchi et al., 2005; Schwab et al., 2006). A more recent study (French, 2004) in which cows were fed a large dose of niacin (45 g/day) for 30 days prior to calving resulted in a 65% decrease in plasma NEFA at calving. This demonstrated that niacin might have a dramatic effect on lipid mobilization if small amounts reach the intestine. Unfortunately, a follow-up study (Chamberlain and French, 2006) could not confirm the results, which may have been related to time of blood sampling.

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20

Recent studies in our laboratory (Pires and Grummer, 2007) have demonstrated that a single dose of 6, 30, or 60 mg niacin/kg bodyweight (BW) to the abomasums of feed-restricted non-lactating, non-gestating cows resulted in a 60–90% reduction in plasma NEFA for 1–4 h followed by a dramatic rebound in plasma NEFA that was 2–3 times higher than pre-dose concentration (from approximately 600 lEq/L to 1200 to 1800 lEq/L). The rebound lasted several hours. A follow-up experiment indicated that the depression in plasma NEFA could be maintained by hourly administration of 6 mg niacin/kg BW (lower doses were not examined). Collectively, these data suggest that high doses of niacin or more modest doses of ruminallyprotected niacin may be potent supplements for control of plasma NEFAs and liver TGs if flow to the intestine is maintained. However, more research is needed. For example, if feed intake of a cow is markedly reduced near or on the day of calving, niacin supplementation could exacerbate the problem of elevated NEFA and fatty liver. Chromium is an essential nutrient for animals. It acts to potentiate the action of insulin as part of the glucose tolerance factor. Since insulin is antilipolytic and its action during the periparturient period may be diminished due to insulin resistance, Cr may have potential to moderate plasma NEFA concentrations and reduce hepatic TG accumulation. Inorganic Cr sources are poorly absorbed; organic forms are more available. We fed 0, 0.03, 0.06, or 0.12 mg of Cr as Cr-methione per day/kg BW0.75 to multiparous cows (12/treatment) from 28 days prior to calving to 28 days post-calving (Hayirli et al., 2001). Pre- and postpartum DMI increased linearly and milk yield increased quadratically with Cr supplementation. Milk composition was not affected. NEFA concentration was decreased linearly pre-partum (but not post-partum) with Cr supplementation. Liver TG content was not affected. Organic Cr supplementation has reduced liver TG in steers (Besong et al., 1996) and decreased blood cortisol, NEFA and beta-hydroxybutyrate in primiparous but not multiparous cows (Subiyatno et al., 1996; Yang et al., 1996). Suppression of milk fat synthesis has been suggested as a mechanism to improve energy balance of fresh cows and, therefore, indirectly reduce adipose tissue lipolysis. Specifically, fatty acids with a trans-10 double bond, particularly trans-10, cis-12 CLA are known to inhibit mammary lipid synthesis and have been examined. In several studies, feeding these compounds to transition cows has not lowered blood NEFA or hepatic TG (Bernal-Santos et al., 2003; Selberg et al., 2004, 2005; Castaneda-Gutierrez et al., 2005; Onetti et al., 2006). The lack of a response in these studies may have been due to an insufficient dose of CLA because mammary lipid synthesis during the first 3 weeks after calving seems to be less sensitive to CLA than later in lactation (Bernal-Santos et al., 2003; Castaneda-Gutierrez et al., 2005). In recent studies where a higher dose of CLA was fed to transition dairy cows and mammary lipid synthesis was affected, hepatic TG was not measured (Moore et al., 2004; Odens et al., 2007).

15

Enhancing hepatic VLDL secretion Choline and methionine are feed additives that have been examined because of their potential to enhance VLDL export from the liver. It is well established that choline deficiency in rats causes an increase in accumulation of TGs in liver. As a methyl donor for biochemical reactions, choline may spare the requirement for methionine (also a methyl donor). Choline also serves as a substrate for synthesis of phosphatidylcholine (PC), a constituent of VLDL. Methionine is an amino acid that is required for synthesis of protein (a constituent of VLDL) and as a methyl donor for PC synthesis. Therefore, if flow of choline to the intestine of dairy cattle is insufficient during the periparturient period when feed intake is low, synthesis of VLDL could be limited and fatty liver could result. The microbial population in the rumen quickly degrades dietary choline. Therefore, the only effective way to assess if ruminants are deficient is to supplement choline postruminally or feed choline in a form that is protected from ruminal degradation (Atkins et al., 1998). Hartwell et al. (2000) fed ruminally-protected choline to transition dairy cows but did not see any beneficial effect on liver TG concentration. However, the degree of ruminal protection of the choline fed in that trial has been questioned. More recently, protected choline was fed to transition dairy cows and a statistically non-significant reduction in liver TG was observed as level of supplementation was increased (0, 11.25, 15, and 18.75 g choline/day; Piepenbrink and Overton, 2004). Similarly, Zahra et al. (2006) did not observe an effect of feeding 14 g choline/day in a protected form on liver TG when initiated 3 weeks prior to calving. Although their study had large cow numbers (n = 182), the standard error for liver TG measurement was quite large. Liver TG is a highly variable measurement in post-partum dairy cattle. Therefore, we have attempted to assess whether choline had a role in preventing or alleviating fatty liver using an experimental model that might be more sensitive for detecting a treatment effect (Cooke et al., 2007). Far-off dry cows were restricted to 30% of maintenance requirements for energy for 10 days to induce fatty liver. During feed restriction, cows received 0 or 15 g of choline/day in a ruminally-protected form. Unexpectedly, choline supplementation significantly reduced plasma NEFA concentration. Likewise, liver TG was significantly reduced; 16.7 and 9.3 mg/mg DNA for control and choline-supplemented cows, respectively, at the end of feed restriction. Therefore, choline reduced TG accumulation, but it is not known if it was a direct effect on the liver or an indirect effect due to reduction in plasma NEFA. In a second experiment (Cooke et al., 2007), cows were similarly feed restricted to induce fatty liver, but choline was not supplemented until after feed restriction when cows resumed ad libitum feed intake. This study indicated that choline was also effective in treating fatty liver because the rate of TG depletion from the liver was increased in choline-supplemented cows after a return to positive

16

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20

energy balance. Choline did not affect plasma NEFA in this study; therefore, the effects appeared to be directly on the liver. In support of our results, transition cows fed 25 g choline/day for 21 days pre-partum and 50 g/day for 60 days post-partum in a protected form had significantly lower liver total lipid at day 7 (146 versus 82 g/kg wet weight) and 35 (85 versus 50 g/kg wet weight) than controls (Elek et al., 2004). Choline that is adequately protected from ruminal degradation may be an effective feed additive for preventing and treating fatty liver. The same model that we used for evaluation of choline (Cooke et al., 2007) was used to evaluate if the methionine hydroxy analog could serve as a lipotropic agent; no effects were detected (Bertics and Grummer, 1999). Bauchart et al. (1998) fed protected methionine or protected methionine plus lysine to cows from 7 to 28 days post-calving and observed a reduction in liver TG with the latter treatment. At this time, there is insufficient evidence to recommend feeding supplemental methionine to moderate TG accumulation in the liver. Altering hepatic fatty acid metabolism It is well known that fat, specifically fatty acids, are more than an energy source and are potent modifiers of metabolism. For example, omega-3 fatty acids are potent regulators of lipid metabolism in liver of rodents and humans. Specifically, they seem to decrease fatty acid conversion to TG and ketones and promote fatty acid oxidation to carbon dioxide. Therefore, we speculated that delivery of certain fatty acids to the liver might prevent TG accumulation in transition cows. Our first experiments (Mashek and Grummer, 2003) examined the effects of purified fatty acids on lipid metabolism of liver cells in culture. From these experiments, it appeared that C18:3 might have beneficial effects and C20:5 and C22:6 may have detrimental effects on liver metabolism. To test this, we subjected dry cows to feed restriction to cause fatty liver while simultaneously infusing intravenously either emulsions of tallow (control, representing fatty acids mobilized from adipose), linseed oil (high in C18:3), or fish oil (high in C20:5 and C22:6) (Mashek et al., 2005). Relative to tallow infusion (12 lg TG/lg DNA), linseed oil (7.8 lg TG/lg DNA) but not fish oil (14.1 lg TG/lg DNA) infusion reduced liver TG accumulation during fatty liver induction. These data supported the cell culture data in indicating that C18:3 may be a useful fatty acid for preventing fatty liver. However, since plasma NEFA concentration was also decreased by linseed oil infusion, we do not know if the effects were direct on the liver or indirect through reducing plasma NEFA. A similar follow-up study was conducted in which water, tallow, or linseed oil was infused into the abomasum (Kulick et al., 2006). Surprisingly, the opposite effects were observed; tallow was more beneficial than linseed oil in reducing liver TG accumulation during feed restriction.

In this study, there were no effects of treatment on plasma NEFA, so effects may have been directly on the liver. Different results between Mashek et al. (2005) and Kulick et al. (2006) suggest that responses to fatty acids may differ depending on the route of delivery and suggests that absorption of fatty acids may trigger metabolic signals (e.g., gut peptides) that influence liver metabolism. The strategy to reduce TG synthesis in the liver by partitioning fatty acids towards oxidation and away from esterification assumes the liver has the capacity to increase oxidation. Liver tissue has a finite requirement for ATP, the energy currency that results from oxidation. Consequently, if partitioning fatty acids towards oxidation is going to be a feasible strategy to reduce TG accumulation, oxidation must be uncoupled from ATP production. Peroxisomal oxidation differs from mitochondrial oxidation in that it is not linked to ATP production and yields heat. Selberg et al. (2005) examined the effect of feeding trans-octadecenoic acids on mRNA coding for carnitine palmitoyltransferase-1 (CPT1), an enzyme responsible for translocation of fatty acids into the mitochondria for oxidation, and peroxisome proliferators-activated receptor alpha (PPARa) which orchestrates many genes in lipid metabolism including those involved in peroxisomal oxidation. Feeding a mixture of trans 18:1 fatty acid, but not calcium salts of CLA, upregulated hepatic PPARa mRNA; neither affected CPT1 mRNA. However, hepatic TG was not affected by feeding either source of fatty acids. Interestingly, feeding supplemental carnitine may reduce hepatic TG (Drackley et al., unpublished data). Management strategies The effects of herd management strategies on prevention and treatment of fatty liver are largely unknown. It is highly likely that management decisions that affect frequency of diet and group changes, space allotment, social interactions among cows and heifers, heat abatement, bunk space, cow comfort, disease control etc., will influence the prevalence and severity of liver TG accumulation during the transition period. In fact, it is possible that management factors such as these will dwarf the magnitude of impact that nutrition may have on fatty liver. As outlined above, nutritional factors such as energy density of the diet, which one would have expected to have a major influence of liver TG level, appear to have little influence. Since the degree of intake decline at calving is related to severity of fatty liver immediately after calving, management practices that ameliorate the decline may help prevent fatty liver (Grummer et al., 2004). We speculated that diet changes at 3 weeks prior to calving and at calving (and grouping changes that might accompany them) that are part of conventional transition cow management may predispose cows to go off feed at a time when they are inherently prone to reductions in feed intake. Feeding one diet the entire 8-week dry period may help reduce the likelihood of this scenario. However, feeding a transition-type diet

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20

Energy balance (Mcal/day)

15

Prepartum

Postpartum

10 5 0 -10

-8

-6

-4

-2

0

2

4

6

8

10

-5 56 d 28 d 0d

-10

25

Prepartum

Postpartum

DMI (kg/day)

20 15 10 56 d 28 d

5

0d 0 -10 -9

-8

-7

-6

-5

-4

-3

-2

-1

1

2

3

4

5

6

7

8

9

10

Week relative to calving Fig. 4. Dry matter intake of cows with different dry period lengths. Treatments are no planned dry period ( ), 28-day dry period ( ), or 56day dry period ( ).

56

Liver TG (µg/µg DNA)

that is moderate in energy for 8 weeks may lead to overconditioned cows and an increased incidence of metabolic disorders (Rukkwamsuk et al., 1999). Feeding one high fiber diet during the entire dry period may be possible. However, questions persist as to whether a dramatic jump from a high fiber diet to a low fiber diet at calving is best for the cow or the rumen microbes. A compromise strategy may be to shorten the dry period and feed one diet with relatively high energy throughout the dry period. The target energy density for this diet would vary depending on the length of the dry period. In other words, as dry period length decreases, the energy density of the diet could increase because there would be less time to accumulate excess body condition. We conducted a trial in which cows were assigned to a 56, 28, or 0 day dry period (Rastani et al., 2005). For cows receiving the 28 or 0 day dry period, a high concentrate lactation diet (minus buffer) was fed throughout the experiment. Energy balance of cows on these treatments is shown in Fig. 3. The remarkable finding was that cows with a 0 day dry period essentially avoided negative energy balance, and the cows on the 28-day dry period had improved energy balance relative to cows that had the conventional 56 day dry period. Part of this was accounted for by reduced milk yield; cows on 28- and 0-day dry periods produced 92% and 80% as much milk, respectively, as those on the 56-day dry period through the first 70 days post-partum. However, the lower milk yield was not accompanied by lower feed intake. In fact, part of the improvement in energy balance could be accounted for by an increase in feed intake (Fig. 4). In a second study, cows were milked 0 (i.e., dried off), 1 or 4/day during the final 28 days of gestation (Rastani et al., 2007). Again, those cows that did not receive a dry period (1 and 4/day milking) did not experience negative energy balance. We only measured liver TGs in the first study (Fig. 5). Liver TG was reduced for cows on 0 day dry period on the day after calving and was reduced for cows on the 0- and 28-day dry period at 35 days after calving. The reduction in liver TG at calving for cows continuously milked probably reflects the greater DMI as well as the mammary gland being an alternate sink to the liver for

17

28

0

30 25 20 15 10 5 0 -30

1

35

Day relative to calving

Fig. 5. Liver triglyceride (TG) at 30 days pre-partum and 1 and 35 days post-partum for cows receiving no planned dry period, 28-day dry period, or 56-day dry period.

the uptake of NEFAs. The advantages in liver TG at 35 days post-partum for cows receiving 0- or 28-day dry period probably represents the more favorable energy balance although blood NEFA concentration was significantly reduced only for the cows on the 0 day dry period. Continuous milking may provide an avenue to effectively reduce fatty liver at calving. The loss in milk production is a major drawback that will prevent the management strategy from being adopted. However, it should be noted that in the second study there was a parity by treatment interaction; cows that were milked 4/day pre-partum did not have a loss in milk production during the subsequent lactation relative to those that had a 28- day dry period, but only if they were in their third or greater lactation. Similarly, in a recent large-scale study on a commercial dairy farm, we observed a parity by treatment interaction when cows were given a 55- versus 34-day dry period (Watters et al., 2006). Third and greater lactation cows did not experience a reduction in milk yield when their dry period was shortened. Further studies must be conducted to determine if there are ways to continuously milk cows and achieve lower hepatic TG without losses in milk yield.

-15

Week relative to calving Fig. 3. Energy balance (Mcal NEl/day) of cows with different dry period lengths. Treatments are no planned dry period ( ), 28-day dry period ( ), or 56-day dry period ( ).

Conclusions During the past 10 years, there have been numerous research trials to examine the effects of diet formulation

18

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20

on hepatic TG accumulation in periparturient cows. Most of these have focused on pre-partum transition diets and have examined the effects of increasing energy density of the diet. It is quite obvious from this literature that increasing energy density of the diet has little effect on liver TG. The one caveat is that more studies need to examine diet formulation strategies during the 3–4 weeks immediately post-partum when negative energy balance is most severe. However, if successful strategies are identified for this period, they will occur too late to avoid the rapid rate of TG accumulation that occurs at calving. Only two diet supplements have been repeatedly proven to reduce TG accumulation in the liver: choline and PG. They probably work by different mechanisms. PG acts to reduce fatty acid mobilization from adipose tissue and choline probably acts to enhance VLDL export from the liver. Hence, they probably could be used in combination to have a greater effect in lowering liver TG. Propylene glycol acts to reduce adipose tissue lipolysis, which can be viewed as an orchestrated metabolic adjustment by the cow to support parturition and lactation. Consequently, it is viewed by some as a less desirable means to control fatty liver. Niacin also has the potential to reduce lipolysis if delivered post-ruminally without degradation in the rumen. However, dramatic increases in plasma NEFAs following cessation of delivery to the intestine may limit its application in periparturient cows that are subject to large fluctuations in feed intake. Improving herd management may be the most effective way to reduce the likelihood of fatty liver; however, that statement is made without sufficient data to support it. Studies to investigate effects of management on fatty liver are logistically difficult to conduct. Results from studies on manipulation of dry period are intriguing and suggest that future studies examining management strategies are warranted. When a cow goes from a dry to a lactating state, energy requirements increase approximately four-fold. In comparison, for a human to increase their energy expenditure four-fold, they would need to jog approximately 6 h/day (Ainsworth et al., 2000). We ask cows to dramatically alter their metabolism with very little time for adjustment, consequently they are susceptible to metabolic disorders such as fatty liver. Continuous milking avoids the sudden metabolic changes that a cow typically experiences during the periparturient period and may relieve metabolic stress. Management strategies that minimize stress on cows need to be evaluated for their effectiveness in preventing or alleviating fatty liver. Conflict of Interest None of the author of this paper has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper.

References Ainsworth, B.E., Haskell, W.L., Whitt, M.C., Irwin, M.L., Swartz, A.M., Strath, S.J., O’Brian, W.L., Basset, D.R., Schmitz, K.H., Emplaincourt, P.O., Jacobs, D.R., Leon, A.S., 2000. Compendium of physical activities: and update of activity codes and MET intensities. Medical Science Sports Exercise 32 (Suppl.), S498–S516. Allen, S.K., Carroll, D.J., Barton, B.A., 1995. The effects of supplemental energy pre- and post-partum on lactation and reproductive performance of dairy cattle. Journal of Dairy Science 78 (Suppl. 1), 164 (Abstract). Atkins, K.B., Erdman, R.A., Vandersall, J.H., 1998. Dietary Choline effects on milk yield and duodenal choline flow in dairy cattle. Journal of Dairy Science 71, 109–116. Bauchart, D., Gruffat, D., Chilliard, Y., 1998. Mechanism of liver steatosis in early lactation cows – effects of hepatoprotector agents. In: Proceedings of the Cornell Nutrition Conference. Cornell University, Ithaca, NY, pp. 27–37. Bernal-Santos, G., Perfield II, J.W., Barbano, D.M., Bauman, D.E., Overton, T.R., 2003. Production responses of dairy cows to dietary supplementation with conjugated linoleic acid (CLA) during the transition period and early lactation. Journal of Dairy Science 86, 3218–3228. Bertics, S.J., Grummer, R.R., 1999. Effects of fat and methionine hydroxy analog on prevention and alleviation of fatty liver induced by feed restriction. Journal of Dairy Science 82, 2731–2736. Besong, S.A., Jackson, J.A., Trammell, D.S., 1996. Influence of supplemental chromium on concentrations of liver triglyceride, blood metabolites and rumen VFA profile in steers fed a moderately high fat diet. Journal of Dairy Science 84, 1679–1685. Boutflour, R.B., 1928. Limiting factors in the feeding and management of milk cows. World Dairy Congress, 15. Castaneda-Gutierrez, Overton, T.R., Butler, W.R., Bauman, D.E., 2005. Dietary supplements of two doses of calcium salts of conjugated linoleic acid during the transition period and early lactation. Journal of Dairy Science 88, 1078–1089. Chamberlain, J.L., French, P.D., 2006. The effects of nicotinic acid supplementation during late gestation on lipolysis and feed intake during the transition period. Journal of Dairy Science 89 (Suppl. 1), 232 (Abstract). Chilliard, Y., 1993. Dietary fat and adipose tissue metabolism in ruminants, pigs, and rodents: a review. Journal of Dairy Science 76, 3897–3931. Christensen, J.O., Grummer, R.R., Rasmussen, F.E., Bertics, S.J., 1997. Method of propylene glycol delivery influences plasma metabolites of feed-restricted cattle. Journal of Dairy Science 80, 563–568. Cooke, R.R., Silva Del Rio, N., Caraviello, D.Z., Bertics, S.J., Ramos, M.H., Grummer, R.R., 2007. Supplemental choline for prevention and alleviation of fatty liver in dairy cattle. Journal of Dairy Science 90, 2413–2418. Dirksen, G., Liebich, H.G., Mayer, E., 1985. Adaptive changes of the ruminal mucosa and their functional and clinical significance. The Bovine Practitioner 20, 116–120. Doepel, L., Lapierre, H., Kennelly, J.J., 2002. Peripartum performance and metabolism of dairy cows in response to pre-partum energy and protein intake. Journal of Dairy Science 85, 2315–2334. Douglas, G.N., Overton, T.R., Bateman II, H.G., Drackley, J.K., 2004. Peripartal metabolism and production of Holstein cows fed diets supplemented with fat during the dry period. Journal of Dairy Science 87, 4210–4220. Duffield, T.F., Sandals, D., Leslie, K.E., Lissemore, K., McBride, B.W., Lumsden, J.H., Dick, P., Bagg, R., 1998. Effect of prepartum administration of monensin in a controlled-release capsule on postpartum energy indicators in lactation dairy cows. Journal of Dairy Science 81, 2354–2361. Elek, P., Husveth, F., Gaal, T., Newbold, J.R., 2004. Effect of rumenprotected choline on liver metabolism in periparturient cows. Journal of Dairy Science 87 (Suppl. 1), 345.

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20 French, P.D., 2004. Nicotinic acid supplemented at a therapeutic level minimizes prepartum feed intake depression in dairy cows. Journal of Dairy Science 87 (Suppl. 1), 345 (Abstract). Friggens, N.C., Andersen, J.B., Larsen, T., Aaes, O., Dewhurst, R.J., 2004. Priming the dairy cow for lactation: a review of dry cow feeding strategies. Animal Research 53, 453–473. Grum, D.E., Drackley, J.K., Younker, R.S., LaCount, D.W., Veenhuizen, J.J., 1996. Nutrition during the dry period and hepatic lipid metabolism of periparturient dairy cows. Journal of Dairy Science 79, 1850– 1864. Grummer, R.R., Carroll, D.J., 1991. Effects of dietary fat on metabolic disorders and reproductive performance of dairy cattle. Journal of Animal Science 69, 3838–3852. Grummer, R.R., 1993. Etiology of lipid-related metabolic disorders in periparturient dairy cows. Journal of Dairy Science 76, 3882–3896. Grummer, R.R., Rastani, R.R., 2003. When should lactating cows reach positive energy balance?. The Professional Animal Scientist 19 197– 203. Grummer, R.R., Mashek, D.G., Hayirli, A., 2004. Dry matter intake and energy balance in the transition period. In: Cook, N.B., Nordlund, K.V. (Eds.), Managing the Transition Cow to Optimize Health and Productivity. Veterinary Clinics of North America. W.B. Saunders Co., Philadelphia, PA, pp. 447–470. Hartwell, J.R., Cecava, M.J., Donkin, S.S., 2000. Impact of dietary rumen undegradable protein and choline on intake, peripartum liver triglyceride, plasma metabolites and milk production in dairy cows. Journal of Dairy Science 83, 2907–2917. Hayirli, A.R., Grummer, R.R., Nordheim, E.V., Crump, P.M., 2002. Animal and dietary factors affecting feed intake during the prefresh transition period in Holsteins. Journal of Dairy Science 85, 3430–3443. Hayirli, A., Bremmer, D.R., Bertics, S.J., Socha, M.T., Grummer, R.R., 2001. Effect of Chromium supplementation on production and metabolic parameters in periparturient dairy cows. Journal of Dairy Science 84, 1218–1230. Holcomb, C.S., Van Horn, H.H., Head, H.H., Hall, M.B., Wilcox, C.J., 2001. Effects of prepartum dry matter intake and forage percentage on postpartum performance of lactating dairy cows. Journal of Dairy Science 84, 2051–2058. Holtenius, K., Agenas, S., Delavaud, C., Chilliard, Y., 2003. Effects of feeding intensity during the dry period. 2. Metabolic and hormonal responses. Journal of Dairy Science 86, 883–891. Ingvartsen, K.L., Aaes, O., Andersen, J.B., 2001. Effects of pattern of concentrate allocation in the dry period and early lactation on feed intake and lactational performance in dairy cows. Livestock Production Science 71, 207–221. Keady, T.W.J., Mayne, C.S., Fitzpatrick, D.A., McCoy, M.A., 2001. Effect of concentrate feed level in late gestation on subsequent milk yield, milk composition, and fertility of dairy cows. Journal of Dairy Science 84, 1468–1479. Kronfeld, D.S., 1982. Major metabolic determinants of milk volume, mammary efficiency, and spontaneous ketosis in dairy cows. Journal of Dairy Science 65, 2204–2212. Kulick, A.E., Greesley, T.F., Pires, J.A.A., Grummer, R.R., 2006. Effects of abomasal lipid infusion on liver triglyceride accumulation during fatty liver induction. Journal of Dairy Science 89 (Suppl. 1), 266 (Abstract). Mashek, D.G., Beede, D.K., 2000. Peripartum responses of dairy cows to partial substitution of corn silage with corn grain in diets fed during the late dry period. Journal of Dairy Science 83, 2310–2318. Mashek, D.G., Grummer, R.R., 2003. Effects of long chain fatty acids on lipid and glucose metabolism in monolayer cultures of bovine hepatocytes. Journal of Dairy Science 86, 2390–2396. Mashek, D.G., Bertics, S.J., Grummer, R.R., 2005. Effects of intravenous triacylglycerol emulsions on hepatic metabolism in fasted dairy cows. Journal of Dairy Science 88, 100–109. Minor, D.J., Trower, S.L., Strang, B.D., Shaver, R.D., Grummer, R.R., 1998. Effects of nonfiber carbohydrate and niacin on periparturient

19

metabolic status and lactation of dairy cows. Journal of Dairy Science 81, 189–200. Moore, C.E., Hafliger, H.C., Mendivil, O.B., Sanders, S.R., Bauman, D.E., Baumgard, L.H., 2004. Increasing amounts of conjugated linoleic acid (CLA) progressively reduces milk fat synthesis immediately postpartum. Journal of Dairy Science 87, 1886–1895. National Research Council, 2001. Nutrient Requirements of Dairy Cattle, seventh ed. Natl. Acad. Press, Washington, DC. Nielsen, N.I., Ingvartsen, K.L., 2004. Propylene glycol for dairy cows – a review of the metabolism of propylene glycol and its effects on physiological parameters, feed intake, milk production and the risk of ketosis. Animal Feed Science Technology 115, 191–213. Odens, L.J., Burgos, R., Innocenti, M., VanBaale, M.J., Baumgard, L.H., 2007. Effects of varying doses of supplemental conjugated linoleic acid on production and energetic variables during the transition period. Journal of Dairy Science 90, 293–305. Onetti, S.G., Bertics, S.J., Luchini, N.D., Grummer, R.R., 2006. Effect of feeding Ca salts of trans-octadecenoic fatty acids and linoleic acid on productive and metabolic responses of dairy cows during the transition period. Journal of Dairy Science 89 (Suppl. 1), 66 (Abstract). Piepenbrink, M.S., Overton, T.R., 2004. Liver metabolism and production of cows fed increasing amounts of rumen-protected choline during the periparturient period. Journal of Dairy Science 86, 1722–1733. Pires, J.A.A., Grummer, R.R., 2007. The use of nicotinic acid to induce sustained low plasma nonesterified fatty acids in feed-restricted Holstein cows. Journal of Dairy Science 90, 3725–3732. Rabelo, E., Rexende, R.L., Bertics, S.J., Grummer, R.R., 2005. Effects of pre- and postfresh transition diets varying in dietary energy density on metabolic status of periparturient dairy cows. Journal of Dairy Science 88, 4375–4383. Rabelo, E., Rezende, R.L., Bertics, S.J., Grummer, R.R., 2003. Effects of transition diets varying in dietary energy density on lactation performance and ruminal parameters of dairy cows. Journal of Dairy Science 86, 916–925. Rastani, R.R., Silva del Rio, N., Dahl, G.E., Grummer, R.R., 2007. Effects of increasing milking frequency during the last 28 days of gestation on milk production, dry matter intake, and energy balance in dairy cows. Journal of Dairy Science 90, 1729–1739. Rastani, R.R., Grummer, R.R., Bertics, S.J., Gu¨men, A., Wiltbank, M.C., Mashek, D.G., Schwab, M.C., 2005. Reducing dry period length to simplify feeding transition cows: milk production, energy balance, and metabolic profiles. Journal of Dairy Science 88, 1004–1014. Reynolds, C.K., Aikman, P.C., Lupoli, B., Humphries, D.J., Beever, D.E., 2003. Splanchnic metabolism of dairy cows through transition from late gestation through early lactation. Journal of Dairy Science 86, 1201–1217. Rukkwamsuk, T., Kruip, T.A., Meijer, G.A.L., Wensing, T., 1999. Hepatic fatty acid composition in periparturient dairy cows with fatty liver induced by intake of a high energy diet in the dry period. Journal of Dairy Science 82, 280–287. Rukkwamsuk, T., Rungruang, S., Choothesa, A., Wensig, T., 2005. Effect of propylene glycol on fatty liver development and hepatic fructose 1,6 bisphoshatase activity in periparturient dairy cows. Livestock Production Science 95, 95–102. Salfer, J.A., Linn, J.G., Otterby, D.E., Hansen, W.P., 1995. Early lactation responses of Holstein cows fed a rumen-inert fat prepartum, postpartum, or both. Journal of Dairy Science 78, 368–377. Santchi, D.E., Berthauiume, R., Matte, J.J., Mustafa, A.F., Girard, C.L., 2005. Fate of supplementary B-vitamins in the gastrointestinal tract of dairy cows. Journal of Dairy Science 88, 2043–2054. Schwab, E.C., Schwab, C.G., Shaver, R.D., Girard, C.L., Putnam, D.E., Whitehouse, N.L., 2006. Dietary forage and nonfiber carbohydrate contents influence B-vitamin intake, duodenal flow, and apparent ruminal synthesis in lactating dairy cows. Journal of Dairy Science 89, 174–187. Selberg, K.T., Lowe, A.C., Staples, C.R., Luchini, N.D., Badinga, L., 2004. Production and metabolic responses of periparturient Holstein

20

R.R. Grummer / The Veterinary Journal 176 (2008) 10–20

cows to dietary conjugated linoleic acid and trans-octadecenoic acids. Journal of Dairy Science 87, 158–168. Selberg, K.T., Staples, C.R., Luchine, N.D., Badinga, L., 2005. Dietary trans octadecenoic acids upregulate the liver gene encoding perosixome-proliferator-activated receptor-a in transition dairy cows. Journal of Dairy Research 72, 107–114. Skaar, T.C., Grummer, R.R., Dentine, M.R., Stauffacher, R.H., 1989. Seasonal effects of pre- and postpartum fat and niacin feeding on lactation performance and lipid metabolism. Journal of Dairy Science 72, 2028–2038. Smith, K.L., Waldron, M.R., Drackley, J.K., Socha, M.T., Overton, T.R., 2005. Performance of dairy cows as affected by prepartum dietary carbohydrate source and supplementation with chromium throughout the transition period. Journal of Dairy Science 88, 255–263. Studer, V.A., Grummer, R.R., Bertics, S.J., Reynolds, C.K., 1993. Effect of prepartum propylene glycol on periparturient fatty liver in dairy cows. Journal of Dairy Science 76, 2931–2939. Subiyatno, A., Mowat, D.N., Yang, Z.W., 1996. Metabolic and hormonal responses to glucose and propionic acid infusions in periparturient

cows supplemented with chromium. Journal of Dairy Science 79, 1436–1445. Vazquez-Anon, M., Bertics, S.J., Luck, M., Grummer, R.R., 1994. Peripartum liver triglyceride and plasma metabolites in dairy cows. Journal of Dairy Science 77, 1521–1528. Waterman, R., Schwalm, J.W., Schultz, L.H., 1972. Nicotinic acid treatment of bovine ketosis I. Effects on circulatory metabolites and interrelationships. Journal of Dairy Science 55, 1447–1453. Watters, R.D., Guenther, J.N., Kulick, A.E., Clark, P.W., Grummer, R.R., 2006. Effect of dry period length on health and production of Holstein cows during the subsequent location. Journal of Dairy Science 89 (Suppl. 1), 213 (Abstract). Yang, W.Z., Mowat, D.N., Subiyatno, A., Liptrap, R.M., 1996. Effects of chromium supplementation on early lactation performance of Holstein cows. Canadian Journal of Dairy Science 76, 221–230. Zahra, L.C., Duffield, T.F., Leslie, K.E., Overton, T.R., Putnam, D., LeBlanc, S.J., 2006. Effects of rumen-protected choline and monensin on milk production and metabolism of periparturient dairy cows. Journal of Dairy Science 89, 4808–4818.