J. Dairy Sci.

88:4384–4392
© American Dairy Science Association, 2005.

Effects of Dietary Cation-Anion Difference on Intake, Milk Yield,

and Blood Components of the Early Lactation Cow
P. S. Chan,1 J. W. West,1 J. K. Bernard,1 and J. M. Fernandez2
1
Department of Animal and Dairy Science, University of Georgia, Coastal Plain Experiment Station, Tifton 31793-0748
2
Department of Animal Sciences, Louisiana State University Agricultural Center, Baton Rouge 70803-4210

ABSTRACT stasis as well as in enzyme and nerve function. Dietary

cation-anion difference may be a superior measure of
Early lactation Holsteins cows (15 primiparous and the effects of electrolytes than the concentration of indi-
18 multiparous) were offered rations with dietary cat- vidual electrolytes provided there are no deficiencies or
ion-anion difference, calculated as mEq (Na + K − Cl − toxicities present. Dietary cation-anion difference has
S)/100 g of feed dry matter (DCAD:S), of 20, 35, or 50 been defined by several equations, including mEq (Na
mEq from d 0 (calving) to 42 d postpartum (August 20, + K − Cl − S) per 100 g of DM (DCAD:S) and mEq (Na
2000 to January 9, 2001) to determine the effects of + K − Cl) per 100 g of DM (DCAD). The DCAD for
increasing DCAD:S on dry matter intake (DMI), milk lactating cows was determined to be more important
yield, and blood metabolites. For DCAD:S of 20, 35, and than individual ingredients (Delaquis and Block,
50, DMI was 3.30, 3.38, 2.96 kg/100 kg of body weight 1995b), cation source (West et al., 1992), or individual
(BW); milk yield was 25.5, 24.2, and 22.4 kg/d, respec- electrolyte concentrations (Tucker and Hogue, 1990);
tively. No differences were observed for concentration and increasing DCAD improved milk yield (Tucker et
or yield of milk fat or milk protein. Serum Ca, P, Mg, al., 1988a; Block, 1994). Early lactation cows are often
Na, K, Cl, cation-anion difference, insulin, and glucose in negative energy balance (Bauman and Currie, 1980)
did not differ with DCAD. Serum HCO3− was 26.07, and may be more susceptible to metabolic acidosis be-
25.88, and 27.64 mEq/L for 20, 35, and 50 DCAD:S. cause of the high concentrate diets fed to meet nutrient
Serum Ca, Mg, Na, and K concentrations were greater needs for lactation. Heat stress, which is often accompa-
for primiparous cows (9.52 mg/dL, 2.35 mg/dL, 140.03 nied by a respiratory alkalosis, may also cause a slight
mEq/L, 4.66 mEq/L, respectively) than for multiparous metabolic acidosis as a compensatory response (Schnei-
cows (9.27 mg/dL, 2.12 mg/dL, 137.63 mEq/L, 4.46 mEq/ der et al., 1988).
L, respectively). A DCAD:S between 23 and 33 mEq/ Higher dietary Na and K concentrations increase the
100 g of dry matter (DM) appears to be adequate during alkalinity of body fluids. The effect of DCAD is mediated
cool weather for the milk yield that occurred in the through a change in acid-base chemistry (Block, 1984)
present study based on DMI (kg/100 kg of BW), whereas and increasing DCAD negates a physiologic acidosis
DCAD:S of 50 mEq/100 g of DM may be excessive and by raising buffering capacity of the blood, reflected by
could be too alkaline or unpalatable, resulting in de- increased blood pH and blood bicarbonate (HCO3−) con-
creased DMI (kg/100 kg of BW). centration. Blood pH and HCO3− concentration in-
(Key words: dietary cation-anion difference, lactation, creased linearly as DCAD:S increased from −10 to 20
dry matter intake, milk yield) mEq (Na + K − Cl − S)/100 g of DM, whereas DMI and
Abbreviation key: CAD = cation-anion difference, milk yield improved by 11 and 9%, respectively, for
DCAD = dietary cation-anion difference, mEq (Na + K DCAD of 20 vs. −10 (Tucker et al., 1988a). Increasing
− Cl) per 100 g of DM, DCAD:S = dietary cation-anion DMI paralleled a linear increase in blood pH for cows
difference, mEq (Na + K − Cl − S) per 100 g of DM. fed rations with DCAD ranging from 12 to 46 mEq (Na
+ K − Cl)/100 g of DM (West et al., 1992). Blood HCO3−
increased from 19 to 28 mEq/L with DCAD:S ranging
INTRODUCTION
from 18 to 52 mEq (Na + K − Cl − S)/100 g of DM, and
Sodium, potassium, and chloride play important roles DMI increased from 20.5 to 24.9 kg/d (E. Block, Arm and
in maintaining osmotic pressure and acid-base homeo- Hammer Animal Nutrition, personal communication).
Cows exposed to hot weather also benefit from greater
DCAD (Escobosa et al., 1984; West et al., 1991, 1992).
Blood HCO3− was maximized with DCAD of 38 mEq
Received April 1, 2005.
Accepted July 18, 2005. (Na + K − Cl)/100 g of DM, whereas DCAD ranging
Corresponding author: Joe W. West; e-mail: [email protected]. from 17 to 38 and 25 to 40 mEq/100 g of DM maximized

4384
 DCAD FOR THE EARLY LACTATION DAIRY COW 4385

DMI and milk yield, respectively (Sanchez et al., 1994). Dwight Co., Inc., Princeton, NJ). Composition of experi-
Intake improved quadratically and milk yield improved mental diets is given in Table 1. Cows were fed a TMR
linearly as DCAD increased from −12 to 31 mEq (West once daily (1300 h) to provide approximately 10% orts
et al., 1991). These 2 studies implied that a higher for ad libitum consumption. Orts were removed daily,
DCAD may be required to support milk yield than for weighed, and daily feed intake was calculated. Ingredi-
DMI. Cows in different stages of lactation may benefit ent proportions were adjusted for DM content each
from differing DCAD (Delaquis and Block, 1995a). week. Cows were milked daily at approximately 0400
Blood pH and HCO3− concentration became more alka- and 1500 h. Milk yield was recorded at each milking
line as the cow progressed from early to late lactation from day of calving through 42 d postpartum. Immedi-
(Erdman et al., 1982). ately following calving, milk was manually weighed in a
The optimal DCAD for prepartum dairy cows is nega- bucket and yield was recorded. After colostrum cleared,
tive (in the range of −5 to −10 mEq/100 g of DM), is milk yield was weighed by electronic meter (Alpro, Alfa
relatively well defined, reduces the incidence of milk Laval Agri, St. Louis, MO). Health-related incidents
fever; and its effectiveness can be monitored via urine were recorded and cows were treated according to
pH (Horst et al., 1997). A positive DCAD improved ADG symptoms.
in poultry (Mongin, 1981; Hulan et al., 1987) and swine
(Patience et al., 1987). When numerous lactation stud- Sampling
ies were summarized to determine the response to
DCAD in terms of DMI and milk yield definite trends Dietary ingredients and TMR were sampled weekly,
were discovered, but the range within which a desirable dried in a forced-air oven at 60°C for 72 h, and DM
response was elicited was relatively broad (Hu and content of TMR was used to calculate DMI for the week.
Murphy, 2004; Sanchez et al., 1994). Thus, the optimal Dried samples were ground to pass through a 6-mm
DCAD for lactating dairy cows has not been well de- screen using a Wiley mill (Arthur Thomas, Philadel-
fined, responses may differ depending on climatic condi- phia, PA), labeled, and stored in sealed plastic bags.
tions (cool or hot weather), and there are few data exam- To improve grinding, whole cottonseed samples were
ining the response of dairy cows to DCAD immediately frozen before and after grinding. Samples were compos-
postpartum. The objectives of this trial were to deter- ited for 3-wk intervals, ground to pass through a 1-mm
mine the effects of increasing DCAD:S on DMI, milk screen using a Wiley mill, and stored for later analysis.
yield, milk composition, serum electrolytes, and acid- Cows were weighed once each week for the duration
base chemistry, and thus, to elucidate the optimal of the study. Weighing was conducted immediately fol-
DCAD:S for the early lactation cow during moderately lowing the p.m. milking and before cows had access to
cool weather. feed and water. Milk samples were collected each week
from 2 consecutive milkings. Blood was collected in
MATERIALS AND METHODS evacuated tubes via jugular venipuncture at 1400 h
during wk 2 and 6 postpartum and placed on ice imme-
Cows and General Management diately following collection. Serum was harvested from
Thirty-three Holstein cows (15 primiparous and 18 blood samples and frozen for later determination of
multiparous) entered the study at parturition. The insulin and glucose.
study began on August 20, 2000, and ended on January
9, 2001. Cows were in the study from calving through Analyses
42 d postpartum. Cows were blocked into groups of 3
according to their 305-d mature equivalent milk for The DM content of feed ingredients and TMR was
the previous lactation (multiparous cows) or estimated determined according to AOAC (1984). Feed N and S
transmitting ability (primiparous cows). Cows were were determined (Leco Corporation, St. Joseph, MI)
housed and fed in a barn with individual free stalls, a and CP was calculated as percentage N × 6.25. Determi-
high metal roof, and open ridge vent, with fans and a nation of ADF was by AOAC (1990) and NDF was by the
high pressure misting system for cow cooling. Cows method of Van Soest et al. (1991). The concentrations of
were trained to use electronic gate feeders (American Na, K, Ca, and Mg were measured by atomic absorption
Calan, Northwood, NH) before calving. Cows within spectrophotometry (model 3030, Perkin-Elmer, Nor-
blocks were assigned randomly to 1 of 3 treatments: walk, CT) following acid digestion (AOAC, 1984). Phos-
DCAD:S of 20, 35, or 50 mEq (Na + K − Cl − S)/100 g phorus was determined colorimetrically using a Beck-
of DM. man DU-60 spectrophotometer (Beckman Instruments
The DCAD:S was formulated using sodium sesquicar- Inc., Fullerton, CA; AOAC, 1984). Chloride was ex-
bonate and potassium carbonate (K-Minus, Church and tracted with a combination of acetic acid and nitric

Journal of Dairy Science Vol. 88, No. 12, 2005

4386 CHAN ET AL.

Table 1. Ingredients and composition of experimental diets.

DCAD:S, mEq/100 g of DM

Item 20 35 50
Ingredient (% of DM)
Corn silage 37.3 36.3 35.1
Alfalfa hay 10.6 10.6 10.6
Whole cottonseed 9.4 9.4 9.4
Steam-flaked corn 22.9 22.8 22.9
Base mix1 15.8 15.9 15.9
Sodium sesquicarbonate2 0.82 1.18 1.58
Potassium carbonate2 0.31 0.99 1.68
Magnesium sulfate2 0.57 0.52 0.52
Soybean meal (48% CP)2 8.7 8.7 8.7
Nutrients (dry basis)
CP, % 16.7 ± 0.3 16.8 ± 0.7 16.2 ± 0.3
Undegradable protein,3 % of CP 38.6 38.7 38.7
NEL,3 Mcal/kg 1.8 1.7 1.7
Ether extract, % 5.0 ± 0.1 4.3 ± 0.6 4.0 ± 0.4
ADF, % 20.7 ± 0.7 20.9 ± 0.5 21.1 ± 0.7
NDF, % 33.9 ± 0.6 33.0 ± 0.6 34.5 ± 0.8
NFC, %3 40.9 40.3 39.8
Ca, % 1.01 ± 0.01 1.00 ± 0.04 1.02 ± 0.03
P, % 0.36 ± 0.02 0.36 ± 0.01 0.35 ± 0.01
Mg, % 0.39 ± 0.06 0.37 ± 0.01 0.37 ± 0.02
K, % 1.29 ± 0.02 1.58 ± 0.06 2.14 ± 0.11
Na, % 0.45 ± 0.01 0.49 ± 0.01 0.63 ± 0.04
Cl, % 0.30 ± 0.02 0.30 ± 0.05 0.29 ± 0.08
S, % 0.33 ± 0.02 0.31 ± 0.01 0.29 ± 0.02
DCAD:S,4 mEq/100 g 22.72 ± 0.62 33.00 ± 2.02 54.64 ± 3.90
1
Contained 46.1% CP, 42.0% undegradable CP, 0.8 Mcal/kg NEL, 11.9% ADF, 16.9% NDF, 19.1% NFC,
3.75% Ca, 1.51% P, 1.22% K, 0.60% Na, 0.91% Cl, 0.49% S, 9696 IU/kg of vitamin A, 3878 IU/kg of vitamin
D, and 48.48 IU/kg of vitamin E.
2
These ingredients were fed as a blended premix.
3
Calculated using NRC (2001) values.
4
Dietary cation-anion difference (DCAD:S) in mEq (Na + K − Cl − S) per 100 g of DM.

acid and was measured using a chloridometer (Haake conducted for a replicated randomized complete block
Buchler Instruments, Inc., Saddle Brook, NJ) by the design with 3 treatments. The model was:
method of Cotlove et al. (1958). Ration and ingredient
fat content was determined by ether extraction (Soxtec Yijkl = ␮ + trti + wkj + park + blkl + (trti × wkj)
System HT6, Tecator, Hoganas, Sweden; AOAC, 1984).
Milk samples were analyzed (Southeast DHIA Labora- + (wkj × park) + (park × blkl) + (trti × park)
tory, McDonough, GA) for fat and protein concentration + (trti × blkl) + (wkj × blkl) + (trti × wkj × park)
by infrared analysis (Multispec; Foss Electric, Hillerød, + (wkj × park × blkl) + (trti × park × blkl) + εijkl
Denmark). Blood serum was analyzed at the University
of Georgia Veterinary Diagnostic Laboratory in Tifton
where ␮ = the mean intercept, trti = the effect due to
for Na, K, Cl, Ca, P, Mg, HCO3− (Roche-Hitachi 912,
treatment i, wkj = the effect due to wk j, park = the
Hoffmann-LaRoche Inc., Indianapolis, IN), and cation-
effect due to parity k, blkl = the effect due to block l,
anion difference (CAD) was calculated using the equa-
tion mEq (Na + K − Cl)/mL. Serum insulin was deter- and ε = error associated with each Yijkl.
mined by radioimmunoassay as described by Bunting Cow within treatment was included as a random ef-
et al. (1994). Serum glucose was determined spectro- fect and week was included as a repeated variable. Sig-
photometrically using a commercial kit (Tech. Bull. 315, nificance was declared at P < 0.05 unless stated other-
Sigma Chemical, St. Louis, MO). wise. Treatment effects were compared using the multi-
ple comparisons approach of Tukey (SAS Institute,
2001). Regression analyses were conducted with the
Statistical Analyses
Proc REG procedure, whereas correlation coefficients
Data were analyzed using the MIXED procedure of were obtained from the Proc CORR procedure of SAS
SAS (SAS Institute, 2001). Analysis of variance was (2001).

Journal of Dairy Science Vol. 88, No. 12, 2005

 DCAD FOR THE EARLY LACTATION DAIRY COW 4387
Table 2. Least square means of production measures for early lactation cows offered DCAD:S of 20, 35, or 50 mEq (Na + K − Cl − S) per
100 g of DM.
P3
DCAD:S,1 mEq/100 g DM Parity2
Trt × Trt ×
Variable 20 35 50 SEM 1 >1 SEM Trt Par Par Wk

DMI, kg/d 16.4 17.1 15.3 0.8 15.6 17.0 0.7 NS NS * NS

DMI, kg/100 kg BW 3.30a 3.38a 2.96b 0.13 3.48 2.94 0.13 * ** * NS
BW, kg 505 509 525 10 456 570 9 NS *** NS *
Milk, kg/d 25.5 24.2 22.4 1.6 21.9 26.1 1.3 NS * † NS
3.5% FCM, kg/d 27.8 25.6 25.7 2.4 23.6 29.1 2.0 NS * NS NS
Fat, % 4.14 3.97 4.31 0.28 4.06 4.22 0.25 NS NS NS *
Protein, % 2.89 2.95 2.91 0.07 2.89 2.94 0.07 NS NS NS NS
Fat, kg 1.05 1.05 1.05 0.17 0.92 1.18 0.10 NS † NS NS
Protein, kg 0.74 0.78 0.69 0.05 0.65 0.82 0.04 NS ** NS NS
Means within a row with different superscripts are different (P < 0.05).
a,b
1
Represents dietary cation-anion difference (DCAD:S) (Na + K − Cl − S).
2
1 represents primiparous cows and >1 represents multiparous cows.
3
Trt = Treatment; Par = Parity.
†P < 0.10; *P < 0.05; **P < 0.01; ***P < 0.0001.

RESULTS AND DISCUSSION study is not common in the literature. The DCAD con-
centrations used by Tucker et al. (1988b) are compara-
Intake and BW
ble with this study; however, only plasma minerals
There were no differences in DMI (kg/d) related to were measured and feed intake and milk yield were
DCAD:S but the DMI (kg/100 kg of BW) was greater not reported in the Tucker study.
(P < 0.05) for diets with DCAD:S of 20 and 35 compared Sanchez et al. (1994) reported a DCAD range from
with the 50 DCAD:S treatment (Table 2). The latter 12 to 62 mEq/100 g of DM and observed optimal DMI
treatment (50 DCAD:S) is equivalent to 73 mEq/100 g for midlactation cows when DCAD was between 30 and
of DM DCAD (S excluded from the equation) and may 50 mEq/100 g of DM. Results for the present study
have caused palatability problems or excess alkalinity, are in agreement with this range, in which DMI was
reducing DMI. Reduced palatability with cationic salts numerically higher for 20 and 35 DCAD:S (equivalent
resulted in depressed feed intake (Stokes et al., 1986). to 43 and 52 mEq/100 g of DM Na + K − Cl) than for
There was a treatment by parity interaction (P < 0.05) 50 DCAD:S. Although the 35 DCAD:S diet is slightly
for DMI (kg/100 kg of BW). Intake for primiparous cows beyond the optimal range suggested by Sanchez et al.
(3.83 kg/100 kg of BW) was greater than that for multip- (1994), the present study was conducted with early lac-
arous cows (2.78 kg/100 kg of BW) at 20 DCAD:S, which tation cows rather than midlactation cows. Early lacta-
was greater than for primiparous and multiparous cows tion cows may require a higher DCAD than cows in
consuming 35 DCAD:S (3.43 and 3.32 kg/100 kg of BW, later lactation (Delaquis and Block, 1995a).
respectively) and 50 DCAD:S (3.19 and 2.73 kg/100 kg
of BW, respectively) treatments. Intake per 100 kg of Milk Yield and Milk Composition
BW for primiparous cows was consistently greater than
for multiparous cows. West et al. (1992) reported that The DCAD:S did not affect milk yield or milk composi-
DMI increased linearly when lactating cows received tion (Table 2), consistent with reports by Roche et al.
DCAD treatments ranging from 12 to 46 mEq/100 g of (2003). Milk yield is dependent on DMI (Kertz et al.,
DM during hot weather. In the present study, 20 and 1991), and in the present study, regression of milk yield
35 DCAD:S were equivalent to DCAD of 43 and 52 mEq on DMI yielded the model y = 4.19 + 1.31x, where y is
(Na + K − Cl)/100 g of DM. milk yield (kg/d) and x is DMI (kg/d) (R2 = 0.54, P <
Our findings are consistent with work by Roche et 0.0001). Lower DMI (kg/100 kg of BW) for 50 DCAD:S
al. (2003), which demonstrated that a DCAD:S above was accompanied by numerically lower milk yield com-
52 mEq/100 g was detrimental to feed intake. Cows pared with 20 and 35 DCAD:S. When DCAD is calcu-
offered the 50 DCAD:S treatment in the present study lated without S in the equation, the 50 DCAD:S treat-
were actually consuming approximately 55 mEq/100 g ment is equivalent to 73 mEq (Na + K − Cl)/100 g of
of DCAD:S (Table 1). The DCAD equation without S DM. Sanchez and Beede (1996) reported that perfor-
was typically used in earlier studies with lactating cows mance was compromised when DCAD was over 50 mEq/
and the very high level of DCAD used in the present 100 g of DM. In contrast, Block (E. Block, Arm and

Journal of Dairy Science Vol. 88, No. 12, 2005

4388 CHAN ET AL.

Hammer Animal Nutrition, personal communication) plementation (Cassida et al., 1988). Multiparous cows
reported improved milk yield up to DCAD:S of 52 mEq/ yielded greater fat and protein than primiparous cows
100 g of DM in a cool-weather study. West et al. (1991) due primarily to greater milk yield. Milk protein yield
reported linear increases in milk yield when DCAD was was positively correlated with DMI (R2 = 0.56, P <
increased from −12 to 31 mEq/100 g of DM. Both milk 0.0001).
yield and 3.5% FCM yield exhibited a parity effect (P
< 0.05), in which production of mature cows was greater Serum Metabolites
than for primiparous cows (Table 2).
Milk yield and DMI were moderate for the present Serum HCO3− was numerically higher for the 50
study (Table 2). Because of the level of DMI, subsequent DCAD:S treatment (Table 3) although statistically, the
ruminal fermentation and organic acid production 20 DCAD:S diet was similar (P < 0.10). Tucker et al.
would be expected to be similarly moderate. It is plausi- (1988b) reported that serum HCO3− was greater for
ble that cows with a higher intake of fermentable carbo- diets containing DCAD:S of 55 vs. 48 mEq/100 g of DM.
hydrates would benefit from a higher DCAD concentra- The 50 DCAD:S diet in the present study may have
tion. A meta-analysis conducted by Hu and Murphy been excessively alkaline as reflected by serum HCO3−
(2004) demonstrated a quadratic milk yield response (27.6 mEq/L), which was at the high end of the physio-
to increasing DCAD. The optimum range was broad, logical range (21.5 to 27.7 mEq/L) described by Benja-
ranging from approximately 25 to 40 mEq/100 g of DM min (1981). An elevation of serum HCO3− may diminish
but average milk yield peaked at about 24 kg/d. Simi- serum Cl concentration because these anions are bal-
larly, DMI peaked near 20 kg/d (again moderate), while anced and exchanged to maintain electroneutrality
blood pH and HCO3− concentration followed patterns (Guyton, 1981). The actual DCAD:S calculated from
similar to DMI (Hu and Murphy, 2004). Sanchez et al. ration analysis for 50 DCAD:S was 54.6 mEq/100 g of
(1994) reported a similar range of DMI and milk yield DM, slightly higher than the DCAD:S 52 mEq/100 g of
response to DCAD, and milk yield peaked at about 22 DM used by Block (E. Block, Arm and Hammer Animal
kg/d. In a summary of dietary buffer studies, Erdman Nutrition, personal communication). However, Block
(1988) reported that cows were responsive to sodium reported improved DMI and milk yield even when blood
bicarbonate supplementation at approximately 30 kg/ HCO3− was 28 mEq/L for the 52 mEq DCAD:S diet.
d milk yield for corn silage-based diets, but were less Similarity among serum HCO3− for 20 and 35
responsive when alfalfa and grass hays and silages DCAD:S treatments paralleled DMI (kg/100 kg of BW;
were fed. Diets supplemented with sodium bicarbonate Table 2. The analyzed DCAD:S for the 20 and 35
would have a greater DCAD value, whereas the diets DCAD:S treatments were 23 and 33 mEq/100 g of DM,
containing alfalfa and grass hay and silage would have and the range was narrower than anticipated. Diets
higher initial DCAD values and would be expected to fed to lactating dairy cows without and with buffer
be less responsive to greater cation content in the diet. supplementation (DCAD concentrations of 27 and 33
Roche et al. (2003) used a wide DCAD range that ex- mEq/100 g of DM, respectively) resulted in no change
tended to very high DCAD values (+21, +52, +102, and in feed intake or milk yield (McKinnon et al., 1990).
+127 mEq/100 g of DM). Those authors reported no Sanchez et al. (1997) reported no difference for DMI,
improvement in milk yield, a small but linear decline milk yield, or milk fat composition when DCAD ranged
in DMI, and a reduced daily weight gain with the rela- from 25 to 40 mEq/100 g of DM. Reports suggest that
tively high and increasing DCAD. Reported milk yield DCAD effects are mediated through blood buffering
was in the range of 23.2 to 25.4 kg/d and DMI varied (West et al., 1992; Delaquis and Block, 1995b). The
around 15 kg/d, and was perhaps inadequate to chal- present trial agrees with the hypothesis because DMI
lenge the systemic buffering capacity of the cow. Fur- (Table 2) and serum HCO3− (Table 3) were the only
ther DCAD research with high-producing dairy cows variables that responded to DCAD:S. Serum HCO3− de-
will be necessary to define the range necessary to pro- clined (P = 0.03) from wk 2 to 6 postpartum, possibly
vide ruminal and systemic buffering for cows consum- due to the increasing DMI over time. Greater consump-
ing large quantities of DM. tion of concentrates increases systemic acidity and de-
Milk fat and protein percentage and yield were not creases serum bicarbonate.
affected by DCAD (Table 2), agreeing with West et al. A treatment by week interaction occurred for serum
(1992) and Delaquis and Block (1995a). Milk fat per- insulin (Table 3). The serum insulin concentration for
centage was relatively high, probably reflecting mobi- the 35 DCAD:S treatment at wk 2 postpartum was
lized adipose tissue in early lactation. Milk yield and lower (P < 0.10) than that for wk 6 postpartum. The
milk protein percentage and yield for early lactation reason for this difference is not apparent. However,
cows were not changed by sodium sesquicarbonate sup- insulin concentrations are positively correlated with

Journal of Dairy Science Vol. 88, No. 12, 2005

 DCAD FOR THE EARLY LACTATION DAIRY COW 4389
Table 3. The effects of DCAD:S on serum metabolites for early lactation cows offered DCAD:S of 20, 35 or 50 mEq (Na + K − Cl − S) per
100 g DM.
P4
DCAD:S2, mEq/100 g DM Parity3
Trt × Trt ×
Variable Week1 20 35 50 Mean SEM 1 >1 SEM Trt Par Par Wk

HCO3−, mEq/L 2 26.6 26.2 28.7 27.2c

6 25.6 25.5 26.6 25.9d
Mean 26.1ab 25.9a 27.6b 0.60 26.17 26.89 0.49 † NS NS NS
Insulin, ␮IU/mL 2 9.4 6.0 9.8 8.4c
6 11.1 14.1 10.2 11.8d
Mean 10.3 10.1 10.0 1.77 10.25 9.94 1.54 NS NS NS †
Glucose, mmol/L 2 3.37 3.00 3.23 3.20c
6 3.84 3.62 3.46 3.64d
Mean 3.60 3.31 3.34 0.14 3.39 3.44 0.11 NS NS NS NS
Serum CAD,5 mEq/L 2 46.5 46.0 48.9 47.1
6 45.6 46.0 46.2 45.9
Mean 46.1 46.0 47.5 0.69 46.78 46.24 0.57 NS NS NS NS
Means within a row with different superscripts are different (P < 0.05).
a,b

Means within a column for each variable with different superscripts are different (P < 0.05).
c,d

1
Week = Week relative to parturition (2) = 2 wk postpartum (6) = 6 wk postpartum.
2
Represents dietary cation-anion difference (Na + K − Cl − S).
3
1 represents primiparous cows and >1 represents multiparous cows.
4
Trt = Treatment; Par = Parity.
5
CAD = Cation-anion difference (Na + K − Cl) in mEq/L.
†P < 0.10.

energy balance (Arieli et al., 2001), and body reserves trations within the normal range (Guyton, 1981), and
are mobilized during negative energy balance (Bauman DCAD had no effect on plasma P, Mg, Na, and K in the
and Currie, 1980). Rastani et al. (2001) reported that work by Tucker et al. (1991). In the present study,
a nadir of tissue energy balance occurred at wk 2 post- serum Mg was greater (P = 0.03) for primiparous cows
partum compared with wk 4 prepartum through wk 17 (2.35 mg/dL) than for mature cows (2.12 mg/dL) and at
postpartum. Because serum insulin in the current wk 6 compared with wk 2 postpartum (Table 4), perhaps
study was lower for wk 2 (8.4 ␮IU/mL) relative to wk because of greater DMI (kg/100 kg of BW) for primipa-
6 (11.8 ␮IU/mL) postpartum (Table 3), energy balance rous cows (Table 2) and greater DMI for wk 6 compared
for cows at wk 2 postpartum was probably more nega- with wk 2 postpartum. Serum Cl concentrations follow
tive than at wk 6 postpartum. serum Na and K concentrations because of the need
There was no effect of DCAD:S on serum CAD, calcu- to maintain electroneutrality (Rose, 1984). Regression
lated as mEq (Na + K − Cl)/mL. However, serum HCO3− analyses for serum Na on Cl and serum K on Cl yielded
had a significant (P < 0.0001) positive relationship with R2 values of 0.51 and 0.24, respectively (P < 0.0001). A
serum CAD (Figure 1). Tucker and Hogue (1990) re- one-unit (1.0 mEq/L) increase in serum Cl concentra-
ported a positive linear relationship between blood tion was accompanied by 0.7 mEq/L increase in serum
HCO3− and serum CAD. Plasma CAD increased signifi- Na and 0.04 mEq/L increase in serum K. The relation-
cantly from 44 to 49 mEq/L for cows fed diets with ship is stronger with Na because Cl is found in the
DCAD:S of 9 vs. 22 mEq/100 g of DM (Tucker et al., extracellular compartment with Na (Guyton, 1981). Se-
1991). Because DCAD affects blood acid-base chemistry rum Cl had a reciprocal relationship with serum HCO3−
as reflected by serum CAD and serum HCO3−, nutrient (r = −0.55, P < 0.0001), which has been observed by
manipulation can be useful in modifying the acid-base others (Escobosa et al., 1984; West et al., 1992; Sanchez
chemistry of cows to improve productivity. Serum CAD et al., 1994).
should be further investigated to determine the optimal
concentration for performance enhancement. CONCLUSIONS
No DCAD:S treatment effects were observed for se-
rum mineral concentrations (Table 4). These results The DMI (kg/100 kg of BW) was similar for 20 and
agree with previous reports (Tucker et al., 1988a; West 35 DCAD:S treatments, which were greater compared
et al., 1992; Delaquis and Block, 1995a). Homeostatic with the 50 DCAD:S treatment. Serum HCO3− tended
mechanisms typically maintain blood Ca and P concen- to be higher for the 50 DCAD:S diet and the concentra-

Journal of Dairy Science Vol. 88, No. 12, 2005

4390 CHAN ET AL.

Figure 1. Simple linear regression of serum HCO3− on serum cation-anion difference (Na + K − Cl, mEq/L) for cows offered dietary
cation-anion difference (DCAD:S) of 20, 35, and 50 mEq (Na + K − Cl − S)/100 g of DM.

tions of serum HCO3− and CAD were positively corre- yield and DMI encountered in the present study, the
lated. The 20 and 35 DCAD:S treatments appeared ade- cows’ acid-base balance may not have been challenged
quate for the early lactation cow during cool weather; sufficiently to benefit from the highest DCAD treat-
50 DCAD:S may have been too alkaline or was unpalat- ment. However, additional work is needed to evaluate
able. In addition, in the presence of the moderate milk the response of high-yielding, early lactation dairy cows

Table 4. The effects of DCAD:S on serum electrolytes for early lactation cows offered DCAD:S of 20, 35, or 50 mEq (Na + K − Cl − S) per
100 g of DM.
P4
DCAD:S,2 mEq/100 g DM Parity3
Trt × Trt ×
Variable Week1 20 35 50 Mean SEM 1 >1 SEM Trt Par Par Wk

Ca, mg/dL 2 9.56 9.16 9.25 9.32

6 9.48 9.48 9.43 9.46
Mean 9.52 9.32 9.34 0.11 9.52 9.27 0.09 NS † * NS
P, mg/dL 2 6.07 6.44 5.80 6.10
6 6.13 6.60 6.38 6.37
Mean 6.10 6.52 6.09 0.26 6.37 6.10 0.21 NS NS NS NS
Mg, mg/dL 2 2.22 2.16 2.12 2.17a
6 2.34 2.31 2.29 2.31b
Mean 2.28 2.24 2.21 0.13 2.35 2.12 0.07 NS * NS NS
Na, mEq/L 2 138.8 140.1 140.7 139.9a
6 135.6 138.9 138.8 137.8b
Mean 137.2 139.5 139.8 0.94 140.03 137.63 0.77 NS * NS NS
K, mEq/L 2 4.33 4.63 4.39 4.45a
6 4.67 4.63 4.72 4.67b
Mean 4.50 4.63 4.55 0.09 4.66 4.46 0.07 NS † NS NS
Cl, mEq/L 2 96.7 98.8 96.2 97.2
6 94.7 97.6 97.3 96.6
Mean 95.7 98.2 96.8 1.14 97.91 95.86 0.93 NS NS NS NS
Means within column for each variable with different superscripts are different (P < 0.05).
a,b
1
Week = Week relative to parturition (2) = 2 wk postpartum (6) = 6 wk postpartum.
2
Represents dietary cation-anion difference (Na + K − Cl − S).
3
1 represents primiparous cows and >1 represents multiparous cows.
4
Trt = Treatment; Par = Parity.
†P < 0.10; *P < 0.05.

Journal of Dairy Science Vol. 88, No. 12, 2005

 DCAD FOR THE EARLY LACTATION DAIRY COW 4391

to DCAD to better define both the DCAD range and the dairy cows: Effects on production, acid-base metabolism and di-
gestion. J. Dairy Sci. 65:712–731.
benefits to cows consuming very high levels of feed DM, Escobosa, A., C. E. Coppock, L. D. Rowe, Jr., W. L. Jenkins, and C.
and for high-producing cows exposed to hot weather. E. Gates. 1984. Effects of dietary sodium bicarbonate and calcium
The DCAD:S may be a better measure of macromineral chloride on physiological responses of lactating dairy cows in hot
weather. J. Dairy Sci. 67:574–584.
requirements rather than individual concentrations,
Guyton, A. C. 1981. Textbook of Medical Physiology. W. B. Saunders
and further studies using DCAD:S in early lactation Co., Philadelphia, PA.
cows should categorize serum CAD as it correlates with Horst, R. L., J. P. Goff, T. A. Reinhardt, and D. R. Buxton. 1997.
serum bicarbonate in cool and hot weather, because Strategies for preventing milk fever in dairy cattle. J. Dairy Sci.
80:1269–1280.
blood bicarbonate can be influenced by environment. Hu, W., and M. R. Murphy. 2004. Dietary cation-anion difference
There may be an optimal serum CAD or HCO3− concen- effects on performance and acid-base status of lactating dairy
tration for the lactating dairy cow. cows: A meta-analysis. J. Dairy Sci. 87:2222–2229.
Hulan, H. W., P. C. M. Simons, P. J. W. Van Schagen, K. B. McRae,
and F. G. Proudfoot. 1987. Effect of dietary cation-anion balance
ACKNOWLEDGMENTS and calcium content on general performance and incidence of leg
abnormalities of broiler chickens. Can. J. Anim. Sci. 67:165–177.
The Church and Dwight Company, Inc. is gratefully Kertz, A. F., L. F. Reutzel, and G. M. Thomson. 1991. Dry matter
acknowledged for furnishing sodium sesquicarbonate intake from parturition to midlactation. J. Dairy Sci. 74:2290–
2295.
and potassium carbonate used in the trial. The authors
McKinnon, J. J., D. A. Christensen, and B. Laarveld. 1990. The influ-
would like to thank Sue Trammell, Heath Cross, Cam- ence of bicarbonate buffers on milk production and acid-base
eron Asbell, Cal Williamson, student workers, and em- balance in lactating dairy cows. Can. J. Anim. Sci. 70:875–886.
ployees from the University of Georgia Dairy Research Mongin, P. 1981. Recent advances in dietary cation-anion balance:
applications in poultry. Proc. Nutr. Soc. 40:285–294.
Center in Tifton for help in feeding and caring for cows National Research Council. 2001. Nutrient Requirements of Dairy
on the trial. Gratitude also goes to Anita Merrill from Cattle. 7th rev. ed. Natl. Acad. Sci., Washington, DC.
the University of Georgia Veterinary Diagnostic Labo- Patience, J. F., R. E. Austic, and R. D. Boyd. 1987. Effect of dietary
ratory for blood analyses. electrolyte balance on growth and acid-base status in swine. J.
Anim. Sci. 64:457–466.
Rastani, R. R., S. M. Andrew, S. A. Zinn, and C. J. Sniffen. 2001.
REFERENCES Body composition and estimated energy balance in Jersey and
Holstein cows during early lactation. J. Dairy Sci. 84:1201–1209.
AOAC. 1984. Official Methods of Analysis. I. 14th ed. Association of Roche, J. R., D. Dalley, P. Moate, C. Grainger, M. Rath, and F.
Official Analytical Chemists International, Arlington, VA. O’Mara. 2003. Dietary cation-anion difference and the health
Arieli, A., S. Abramson, S. J. Mabjeesh, S. Zamwel, and I. Bruckental. and production of pasture-fed dairy cows. 1. Dairy cows in early
2001. Effect of site and source of energy supplementation on milk lactation. J. Dairy Sci. 86:970–978.
yield in dairy cows. J. Dairy Sci. 84:462−470. Rose, B. D. 1984. Clinical Physiology of Acid-Base and Electrolyte
Bauman, D. E., and W. B. Currie. 1980. Partitioning of nutrients Disorders. 2nd ed. McGraw-Hill Book Co., New York, NY.
during pregnancy and lactation: A review of mechanisms involv- Sanchez, W. K., and D. K. Beede. 1996. Is there an optimal cation-
ing homeostasis and homeorhesis metabolism of dairy cows. J.
anion difference for lactation diets? Anim. Feed Sci. Technol.
Dairy Sci. 63:1514−1529.
59:3–12.
Benjamin, M. M. 1981. Fluid and electrolytes. Page 213 in Outline
of Veterinary Clinical Pathology. Iowa State Univ. Press, Ames. Sanchez, W. K., D. K. Beede, and J. A. Cornell. 1994. Interactions
Block, E. 1984. Manipulating dietary anions and cations for prepar- of sodium, potassium and chloride on lactation, acid-base status
tum dairy cows to reduce incidence of milk fever. J. Dairy Sci. and mineral concentrations. J. Dairy Sci. 77:1661–1675.
67:2939−2948. Sanchez, W. K., D. K. Beede, and J. A. Cornell. 1997. Dietary mixtures
Block, E. 1994. Manipulation of dietary cation-anion difference on of sodium bicarbonate, sodium chloride, and potassium chloride:
nutritionally related production diseases, productivity, and meta- Effects on lactational performance, acid-base status, and mineral
bolic responses of dairy cows. J. Dairy Sci. 77:1437–1450. metabolism of Holstein cows. J. Dairy Sci. 80:1207–1216.
Bunting, L. D., J. M. Fernandez, D. L. Thompson, Jr., and L. L. SAS Institute. 2001. SAS User’s Guide. Statistics, Version 8 ed. SAS
Southern. 1994. Influence of chromium picolinate on glucose us- Inst., Inc., Cary, NC.
age and metabolic criteria in growing Holstein calves. J. Anim. Schneider, P. L., D. K. Beede, and C. J. Wilcox. 1988. Nycterohemeral
Sci. 72:1591–1599. patterns of acid-base status, mineral concentrations and digestive
Cassida, K. A., L. D. Muller, and T. F. Sweeney. 1988. Sodium sesqui- function of lactating cows in natural or chamber heat stress envi-
carbonate for early lactation dairy cows fed corn silage-based ronments. J. Anim. Sci. 66:112–125.
diets. J. Dairy Sci. 71:381–387. Stokes, M. R., L. L. Vandemark, and L. S. Bull. 1986. Effects of
Cotlove, E. H., H. V. Trantham, and R. L. Bowman. 1958. An instru- sodium bicarbonate, magnesium oxide, and a commercial buffer
ment and method for automatic, accurate, and sensitive titration mixture in early lactation cows fed hay crop silage. J. Dairy Sci.
of chloride in biological samples. J. Lab. Clin. Med. 51:461–465.
69:1595–1603.
Delaquis, A. M., and E. Block. 1995a. Dietary cation-anion difference,
Tucker, W. B., G. A. Harrison, and R. W. Hemken. 1988a. Influence
acid-base status, mineral metabolism, renal function and milk
production of lactating cows. J. Dairy Sci. 78:2259–2284. of dietary cation-anion balance on milk, blood, urine and rumen
Delaquis, A. M., and E. Block. 1995b. The effects of changing ration fluid in lactating dairy cattle. J. Dairy Sci. 71:346–354.
ingredients on acid-base status, renal function, and macromineral Tucker, W. B., and J. F. Hogue. 1990. Influence of sodium chloride
metabolism. J. Dairy Sci. 78:2024–2039. or potassium chloride on systemic acid-base status, milk yield
Erdman, R. A. 1988. Dietary buffering requirements of the lactating and mineral metabolism in lactating dairy cows. J. Dairy Sci.
dairy cow: A review. J. Dairy Sci. 71:3246–3266. 73:3485–3493.
Erdman, R. A., R. W. Hemken, and L. S. Bull. 1982. Dietary sodium Tucker, W. B., J. F. Hogue, D. F. Waterman, T. S. Swenson, Z. Xin,
bicarbonate and magnesium oxide for early postpartum lactating R. W. Hemken, J. A. Jackson, G. D. Adams, and L. J. Spicer. 1991.

Journal of Dairy Science Vol. 88, No. 12, 2005

4392 CHAN ET AL.

Role of sulfur and chloride in the dietary cation-anion balance charides in relation to animal nutrition. J. Dairy Sci. 74:3583–
equation for lactating dairy cattle. J. Anim. Sci. 69:1205–1213. 3597.
Tucker, W. B., Z. Xin, and R. W. Hemken. 1988b. Influence of dietary West, J. W., K. D. Haydon, B. G. Mullinix, and T. G. Sandifer. 1992.
calcium chloride on adaptive changes in acid-base status and Dietary cation-anion balance and cation source effects on produc-
tion and acid-base status of heat-stressed cows. J. Dairy Sci.
mineral metabolism in lactating dairy cows fed a diet high in 75:2776–2786.
sodium bicarbonate. J. Dairy Sci. 71:1587–1597. West, J. W., B. G. Mullinix, and T. G. Sandifer. 1991. Changing
Van Soest, P. J., J. B. Robertson, and B. A. Lewis. 1991. Methods dietary electrolyte balance for dairy cows in cool and hot environ-
for dietary fiber, neutral detergent fiber, and nonstarch polysac- ments. J. Dairy Sci. 74:1662–1674.

Journal of Dairy Science Vol. 88, No. 12, 2005