Advertisement

Hay to Reduce Dietary Cation-Anion Difference for Dry Dairy Cows

      Abstract

      Timothy grass has a lower dietary cation-anion difference [DCAD = (Na + K) - (Cl + S)] than other cool-season grass species. Growing timothy on low-K soils and fertilizing it with CaCl2 could further decrease its DCAD. The objective of this study was to evaluate the effects of feeding low-DCAD timothy hay on dry dairy cows. Six nonpregnant and nonlactating cows were used in a replicated 3 × 3 Latin square. Treatments were as follows: 1) control diet (control; DCAD = 296 mEq/kg of dry matter); 2) low-DCAD diet based on low-DCAD timothy hay (L-HAY; DCAD = - 24 mEq/kg of dry matter); and 3) low-DCAD diet using HCl (L-HCl; DCAD = - 19 mEq/kg of dry matter). Decreasing DCAD with L-HAY had no effect on dry matter intake (11.8 kg/d) or dry matter digestibility (71.5%). Urine pH decreased from 8.21 to 5.89 when L-HAY was fed instead of the control. Blood parameters that decreased with L-HAY were base excess (− 0.4 vs. 3.8 mM) and HCO3 (23 vs. 27 mM), and blood parameters that increased were Ca2+ (5.3 vs. 5.1 mg/dL), Cl (30.5 vs. 29.5 mg/dL), and Na+ (60.8 vs. 60.1 mg/dL). Compared with the control, L-HAY resulted in more Ca in urine (13.4 vs. 1.2 g/d). Comparing L-HAY with L-HCl, cow dry matter intake tended to be higher (11.5 vs. 9.8 kg/d), and blood pH was higher (7.37 vs. 7.31). Urine pH; total dry matter; Ca, K, P, and Mg apparent absorption; and Ca, K, Na, Cl, S, P, and Mg apparent retention were similar. Absorption as a percentage of intake of Na and Cl was lower for L-HAY as compared with L-HCl. In an EDTA-challenge test, cows fed L-HAY regained their initial level of blood Ca2+ twice as quickly as the control treatment (339 vs. 708 min); there were no differences between L-HAY and L-HCl. This experiment confirms that feeding low-DCAD hay is an effective means of decreasing the DCAD of rations and obtaining a metabolic response in dry dairy cows.

      Key words

      Introduction

      Clinical hypocalcemia, also known as milk fever, is a widespread metabolic disorder that affects about 5.2% of dairy cows at calving (
      USDA
      Part 1: Reference of dairy health and management in the United States, 2002.
      ). This condition can be fatal if not treated, and cows with milk fever are susceptible to other metabolic disorders, including retained fetal placenta, displaced abomasums, and mastitis (
      • Curtis C.R.
      • Erb H.N.
      • Sniffen C.J.
      • Smith R.D.
      • Powers P.A.
      • Smith M.C.
      • White M.E.
      • Hillman R.B.
      • Pearson E.J.
      Association of parturient hypocalcemia with eight periarturient disorders in Holstein cows.
      ;
      • Gröhn Y.T.
      • Erb H.N.
      • McCulloch C.E.
      • Saloniemi H.S.
      Epidemiology of metabolic disorders in dairy cattle: Association among host characteristics, disease, and production.
      ), as well as decreased milk yield (
      • Block E.
      Manipulating dietary anions and cations for prepartum dairy cows to reduce incidence of milk fever.
      ). The subclinical form of milk fever affects 50% of all cows at calving and 66% of those in their third lactation or greater (
      • Beede D.K.
      • Sanchez W.K.
      • Wang C.
      Macrominerals.
      ). Up to 22% are still deficient in Ca 10 d after calving (
      • Goff J.P.
      • Horst R.L.
      • Jardon P.W.
      • Borelli C.
      • Wedam J.
      Field trials of an oral calcium propionate paste as an aid to prevent milk fever in periparturient dairy cows.
      ). Subclinical hypocalcemia may also increase the risk of dairy cows developing other metabolic disorders (
      • Huber T.L.
      • Wilson R.C.
      • Stattelman A.J.
      • Goetsch D.D.
      Effect of hypocalcemia on motility of the ruminant stomach.
      ;
      • Oetzel G.R.
      • Olson J.D.
      • Curtis C.R.
      • Fettman M.J.
      Ammonium chloride and ammonium sulfate for prevention of parturient paresis in dairy cows.
      ).
      Decreasing the DCAD of precalving rations can reduce the incidence of milk fever (
      • Dishington I.W.
      Prevention of milk fever (hypocalcemic paresis puerperalis) by dietary salt supplements.
      ;
      • Charbonneau E.
      • Pellerin D.
      • Oetzel G.R.
      Impact of lowering dietary cation-anion difference in nonlactating dairy cows: A meta-analysis.
      ). Anionic salts (
      • Vagnoni D.B.
      • Oetzel G.R.
      Effects of dietary cation-anion difference on the acid-base status of dry cows.
      ), commercial products (
      • Vagnoni D.B.
      • Oetzel G.R.
      Effects of dietary cation-anion difference on the acid-base status of dry cows.
      ), and HCl (
      • Goff J.P.
      • Horst R.L.
      Use of hydrochloric acid as a source of anions for prevention of milk fever.
      ;
      • Goff J.P.
      • Ruiz R.
      • Horst R.L.
      Relative acidifying activity of anionic salts commonly used to prevent milk fever.
      ) have all proven effective at reducing the DCAD of dry cow diets.
      • Horst R.L.
      • Goff J.P.
      Milk fever and dietary potassium.
      and
      • Horst R.L.
      • Goff J.P.
      • Reinhardt T.A.
      • Buxton D.R.
      Strategies for preventing milk fever in dairy cattle.
      suggested that decreasing the amount of K in forage fed to cows before calving can also prevent hypocalcemia.
      • Tremblay G.F.
      • Brassard H.
      • Bélanger G.
      • Séguin P.
      • Drapeau R.
      • Bregard A.
      • Michaud R.
      • Allard G.
      Dietary cation anion difference of five cool-season grasses.
      , in comparing 5 cool-season grasses, concluded that timothy had the lowest DCAD. Fertilizing with chloride could further decrease timothy hay DCAD (
      • Pelletier S.
      • Bélanger G.
      • Tremblay G.F.
      • Séguin P.
      • Drapeau R.
      • Allard G.
      Dietary cation-anion difference of Timothy (Pleum pratense L.) as influenced by application of chloride and nitrogen fertilizer.
      ).
      There is limited research on the absorption and retention of minerals by dry cows in relation to DCAD. Most studies examine Ca, Mg, and P absorption and retention and do not relate marked differences to DCAD (
      • Wang C.
      • Beede D.K.
      Effects of supplemental protein on acid-base status and calcium metabolism of nonlactating Jersey cows.
      ;
      • Kume S.
      • Toharmat T.
      • Nonaka K.
      • Oshita T.
      • Nakui T.
      • Ternouth J.H.
      Relationship between crude protein and mineral concentrations in alfalfa and value of alfalfa silage as a mineral source for periparturient cows.
      ).
      • Schonewille J.T.
      • Van’t Klooster A.T.
      • Dirkzwager A.
      • Beynen A.C.
      Stimulatory effect of an anion (chloride)-rich ration on apparent calcium absorption in dairy cows.
      evaluated the apparent absorption of Ca, Mg, and P for diets with high (276 mEq/kg) and low DCAD (− 170 mEq/kg) and found an increase in Ca absorption, and a tendency for increased Mg absorption, in cows fed a low-DCAD diet. Two studies (
      • Leclerc H.
      • Block E.
      Effects of reducing dietary cation-anion balance for prepartum dairy cows with specific reference to hypocalcemic parturient paresis.
      ;
      • Delaquis A.M.
      • Block E.
      Acid-base status, renal function, water, and macromineral metabolism of dry cows fed diets differing in cation-anion difference.
      ) reported the apparent absorption and retention of Ca, Mg, P, Na, K, Cl, and S. In those experiments, all diet treatments had a positive DCAD.
      • Leclerc H.
      • Block E.
      Effects of reducing dietary cation-anion balance for prepartum dairy cows with specific reference to hypocalcemic parturient paresis.
      measured the apparent absorption of minerals over a long duration (pre- and postcalving).
      • Delaquis A.M.
      • Block E.
      Acid-base status, renal function, water, and macromineral metabolism of dry cows fed diets differing in cation-anion difference.
      tested a slight variation in DCAD (481 vs. 327 mEq/kg). In both studies, only small differences were observed in apparent absorption and retention of the minerals. Further research is needed to determine how mineral absorption and retention relates to DCAD difference in dry cow rations.
      The objective of our study was to evaluate the effects of a low-DCAD timothy hay diet on DMI, acid-base metabolism, and apparent absorption and retention of minerals in dry dairy cows. Our hypothesis was that timothy hay fertilized with CaCl2 would be as effective as HCl at lowering the DCAD of dry cow rations and preventing hypocalcemia. Dry dairy cows were fed normal- or low-DCAD timothy hay to evaluate the effects of low-DCAD hay on DMI and blood and urine components. The low-DCAD timothy hay treatment was also compared with a positive control treatment that used HCl to decrease DCAD. The apparent digestibility of fiber, N, and minerals, as well as the apparent retention of N and minerals, was evaluated for all 3 treatments.

      Materials and Methods

      Hay Production

      Two types of timothy hay were produced on 2 different fields. Low-DCAD hay was produced on a field with low soil K content (101 kg/ha), and high-DCAD hay was produced on a field with a soil K content of 289 kg/ha. Before the start of spring growth, both fields received 80 kg of N/ha, and, based on the recommendations of
      • Pelletier S.
      • Bélanger G.
      • Tremblay G.F.
      • Séguin P.
      • Drapeau R.
      • Allard G.
      Dietary cation-anion difference of Timothy (Pleum pratense L.) as influenced by application of chloride and nitrogen fertilizer.
      , the field for low-DCAD hay received 140 kg of Cl/ha as CaCl2. Hay from spring growth was cut at the early heading stage and conserved in small bales in a barn equipped with a hay drier.

      Cows, Diets, and Experimental Design

      Six multiparous nonpregnant and nonlactating Holstein cows were used in a replicated 3 × 3 Latin square design with 2-wk periods. All cows were fed, as TMR, chopped timothy hay, ground corn, corn gluten meal, and a mixture of vitamins and minerals. Treatments were (Table 1) as follows: 1) high-DCAD diet (control); 2) low-DCAD diet, using only low-DCAD hay as forage (L-HAY); and 3) low-DCAD diet, using HCl to decrease the DCAD of the control diet (L-HCl). To prevent excessive DMI depression, rations should not exceed a maximal DCAD diminution of 2,300 mEq/d that can be achieved using anionic salts (
      • Oetzel G.R.
      • Barmore J.A.
      Intake of a concentrate mixture containing various anionic salts fed to pregnant, nonlactating dairy cows.
      ). For that reason, low-DCAD hay had to be mixed to high-DCAD timothy hay in the control and L-HCl diets (Table 1); diets formulated with only the high-DCAD timothy hay obtained a DCAD too high to be decreased with HCl alone to the same level as L-HAY. Content differences between hay were also diminished using this mixture, which made the control and L-HAY treatments more comparable in terms of chemical composition (Table 2). Concentrated HCl was diluted with water and molasses (acid:water:molasses; 10:10:4.5, vol/vol/vol) twice a week and added daily to the L-HCl diet. Cows fed control and L-HAY received the same proportion of water and molasses (water:molasses; 10:4.5, vol/vol) in their diet as the cows on L-HCl. Diets were formulated based on
      NRC
      Nutrient Requirements of Dairy Cattle.
      recommendations for transition cows; they all provided a similar level of NEL (1.48 Mcal/kg) and CP (14.6%). Rations were fed once a day, in the morning, to provide 10% orts on an as-fed basis according to the intake of the previous day. Before the experiment began, a 2-wk acclimation period was set aside for cow adaptation to the experimental feeds. The experimental protocol was approved by the Laval University Animal Care Committee, and animals were cared for according to the guidelines of the
      Canadian Council of Animal Care
      .
      Table 1Ingredients of experimental diets
      Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      (%, DM basis)
      ItemControlL-HAYL-HCl
      High-DCAD hay55.5054.9
      Low-DCAD hay23.078.522.7
      Ground corn14.614.614.4
      Corn gluten meal5.25.25.1
      Molasses1.11.11.1
      HCl001.1
      Vitamins and minerals
      Mineral composition: Ca = 16.28%; P = 0.01%; Na = 14.96%; Cl = 23.04%; Mg = 0.88%; K = 0.05%; S = 0.02%; Cu = 993 mg/kg; Mn = 4 mg/kg; Zn = 3,599 mg/kg; Fe = 1,500 mg/kg; Co = 8.28 mg/kg; I = 66.23 mg/kg; Se = 110.08 mg/kg; vitamin A = 1,382 UI/kg; vitamin D = 264 UI/kg; vitamin E = 13,418 UI/kg.
      0.60.60.6
      1 Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      2 Mineral composition: Ca = 16.28%; P = 0.01%; Na = 14.96%; Cl = 23.04%; Mg = 0.88%; K = 0.05%; S = 0.02%; Cu = 993 mg/kg; Mn = 4 mg/kg; Zn = 3,599 mg/kg; Fe = 1,500 mg/kg; Co = 8.28 mg/kg; I = 66.23 mg/kg; Se = 110.08 mg/kg; vitamin A = 1,382 UI/kg; vitamin D = 264 UI/kg; vitamin E = 13,418 UI/kg.
      Table 2Chemical composition of forage and experimental diets
      HayDiet
      Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Chemical componentHigh DCADLow DCADControlL-HAYL-HClSE
      DM, %88.687.585.985.684.30.28
      NEL,
      Estimated for each treatment using the average characteristics of the treatment with NRC (2001) model.
      Mcal/kg
      1.481.481.47
      ADF, % of DM35.935.931.0530.9230.320.438
      NDF, % of DM63.264.657.2856.4853.761.118
      CP, % of DM12.613.514.3015.0514.260.451
      DCAD,
      DCAD (mEq/kg of DM) = (Na + K)-(Cl + S).
      mEq/kg of DM
      5268296− 24− 1921.6
      Na, % of DM0.0040.0140.100.080.100.018
      K, % of DM2.812.162.311.872.240.042
      Cl, % of DM0.281.480.731.381.770.036
      S, % of DM0.190.210.210.240.220.005
      Ca, % of DM0.380.490.470.500.430.022
      P, % of DM0.300.260.310.290.310.006
      Mg, % of DM0.100.160.120.160.130.004
      1 Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      2 Estimated for each treatment using the average characteristics of the treatment with
      NRC
      Nutrient Requirements of Dairy Cattle.
      model.
      3 DCAD (mEq/kg of DM) = (Na + K) - (Cl + S).

      Measurements and Sampling

      Forage, TMR, Orts, and BW

      Feed intake was recorded daily. Forages and TMR were sampled on d 8, 9, and 10 of each period, and orts were sampled the following day. Samples were kept at - 20°C for chemical analysis. Cow BW was recorded twice: at d 11 and 14 of each period.

      Feces

      Total fecal excreted was weighed and sampled twice daily: 10 h postfeeding and at feeding the next morning on d 8, 9, and 10. The first sample was kept at 4°C until the second sample was taken. The 2 samples from the same cow in a 24-h period were pooled in proportion to excretion and were immediately frozen at - 20°C until further analysis.

      Urine

      Bardex catheters (24 mm; 75-mL balloon) were installed into the bladder of each cow on d 7. At feeding on d 8, catheters were connected to a container using polyvinyl chloride tubing. Light mineral oil (50 mL) and 1 g of thymol were added in urine containers to prevent deterioration of urine (
      • Delaquis A.M.
      • Block E.
      Acid-base status, renal function, water, and macromineral metabolism of dry cows fed diets differing in cation-anion difference.
      ). Total collection of urine was done on d 8, 9, and 10. Urine was weighed twice a day, 10 h after feeding and at feeding the next morning. Urine was sampled under light mineral oil through polyvinyl chloride tubing. Volume mass was determined by weighing 1 L of urine. Two 50-mL samples per cow per sampling were immediately frozen at - 20°C until further analysis.

      Blood Samples and Infusion of EDTA

      On d 11 of each experimental period, cows were administered 20 mg of i.v. xylazine tranquilizer (Rompun, Bayer Inc., Toronto, Canada), and catheters were introduced into both jugular veins. Starting at feeding on d 12, hypocalcemia was induced by i.v. infusion of sterile 7% (wt/vol) Na2-EDTA·2H2O (Laboratoire Mat Inc., Beauport, Québec) solution (pH 7.4) at a rate of 0.6 mL of solution/h per kilogram of BW by the means of a peristaltic pump (Micro Macro Plum XL3, Abbott Laboratories, Chicago, IL). The solution was made by mixing 70 g of Na2-EDTA·2H2O in sterilized saline solution (0.9% NaCl) to obtain a final volume of 1 L. Sodium hydroxide solution (5 N) was used to stabilize pH at 7.4 during the process. The solution was sterilized by filtration (0.2 μm). Infusion of Na2-EDTA·2H2O solution was stopped when cow blood Ca2+ had reached approximately half the initial level as determined using an automated microblood gas analyzer (ABL 77, Radiometer, Copenhagen, Denmark). The cows were allowed to recover spontaneously after the infusion, and blood Ca2+ was monitored until the Ca2+ level reached the initial concentration. Complete recovery was considered at 95% of initial blood Ca2+, because, on average, a 5% variation in blood Ca2+ could be observed between samples during baseline measurements. Blood samples were taken at catheter installation, before the infusion, every 10 min during the EDTA infusion and every 30 min after the infusion. Two blood samples were taken to confirm the initial blood Ca2+, the level to stop the infusion, and the total recovery. Blood samples were taken with heparinized syringes balanced for electrolytes (PICO 50, London Scientific Limited, London, Canada) for immediate analysis of blood pH, partial CO2 and O2 pressures, base excess, and whole-blood concentration of HCO3, Na+, K+, Cl, and Ca2+ with an automated microblood gas analyzer (ABL 77, Radiometer). Volume of solution infused, time to decrease Ca2+ to half the initial concentration, and time for complete recovery were monitored during the experiment.

      Chemical Analysis

      Forage, TMR, and Orts

      Forage, TMR, and orts samples were freeze-dried, ground in a Wiley mill (Arthur H. Thomas Co., Philadelphia, PA) through a 1-mm screen, and pooled proportionally to their original quantity by cow by period. Subsamples of ground forages, TMR, and orts were analyzed for ADF and NDF using the Ankom (Ankom200 Fiber Analyzer, Fairport, NY) immersion method (
      • Van Soest P.J.
      • Robertson J.B.
      • Lewis B.A.
      Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition.
      ). Subsamples of ground forage, TMR, and orts were mineralized using a mixture of sulfuric and selenious acid as described by
      • Isaac R.A.
      • Johnson W.C.
      Determination of total nitrogen in plant tissue, using a block digestor.
      . Crude protein concentration was determined using total N (method 13-107-6-2-E;

      Lachat Instruments. 2005; Subject: Method list for automated ion analyzers (flow injection analyses, ion chromatography). http://www.lachatinstruments.com/applications/MethodsList.PDF Accessed April 18, 2005.

      ) that was measured using a Lachat QuikChem 8000 flow injection analysis system (Zellweger Analytics Inc., Lachat Instruments Division, Milwaukee, WI). Total P (method 15-115-01-4-A;

      Lachat Instruments. 2005; Subject: Method list for automated ion analyzers (flow injection analyses, ion chromatography). http://www.lachatinstruments.com/applications/MethodsList.PDF Accessed April 18, 2005.

      ) was measured simultaneously with N. Flame emission was used to determine K concentration. The same extract as N, P, and K was used to determine Mg concentration by atomic absorption spectrometry (Perkin Elmer 3300, Überlingen, Germany). Subsamples of ground forages, TMR, and orts were mineralized at 500°C for 4 h. Ashes were dissolved with 1.0 N HCl (
      • Miller R.O.
      High temperature oxidation.
      ). Concentrations of Na and Ca were determined by atomic absorption spectrometry (Perkin Elmer 3300). Subsamples of ground forages, TMR, and orts were mixed with 20 mL of 0.0007 M sulfuric acid (
      • Liu L.
      Determination of chloride in plant tissue.
      ) for 60 min, centrifuged at 32,570 × g for 30 min, and the Cl concentration of the supernatant was determined by conductivity on a Dionex DX500 equipped with a AS11HC column (Dionex Corporation, Sunnyvale, CA). Subsamples of ground forages, TMR, and orts were digested in nitric acid (
      • Mills H.A.
      • Jones Jr., J.B.
      Plant Analysis Handbook II: A Practical Sampling, Preparation, Analysis, and Interpretation Guide.
      ). Organic and inorganic forms of S were converted to a sulfate form that was precipitated with acidified barium chloride, suspended in a colloidal form, and analyzed by turbidimetry (adaptation of method 10-116-10-1-G;

      Lachat Instruments. 2005; Subject: Method list for automated ion analyzers (flow injection analyses, ion chromatography). http://www.lachatinstruments.com/applications/MethodsList.PDF Accessed April 18, 2005.

      ) on a Lachat QuikChem 8000 flow injection analysis system (Zellweger Analytics Inc., Lachat Instruments Division). Daily intake of ADF, NDF, CP, K, Na, Cl, S, Ca, P, and Mg was calculated by subtracting ort nutrients from TMR nutrients. Ort and TMR nutrients were calculated by multiplying chemical composition with the corresponding amount of orts or TMR.

      Feces

      Fecal samples were freeze-dried. Once dried, they were ground in a Wiley mill (Arthur H. Thomas Co.) through a 1-mm screen. Subsamples were pooled by cow and period, proportionally to daily fecal excretion. Feces were then analyzed for ADF, NDF, CP, K, Na, Cl, S, Ca, P, and Mg, using procedures previously described for forages, TMR, and orts. Daily excretion of ADF, NDF, CP, and macrominerals was then calculated by multiplying fecal excretion by its nutrient concentration. Apparent digestion of ADF, NDF, CP, and the apparent absorption of macrominerals were calculated by subtracting nutrient daily excretion from daily intake.

      Urine

      Urine pH was taken immediately after sampling (Oakton pH 10 Series, Vernon Hills, IL). Thawed subsamples of urine were pooled by cows and period proportionally to the urine excreted. Analysis of N, K, Na, Cl, S, Ca, P, and Mg were then completed without any mineralization, using procedures previously described for forages, TMR, and orts.

      Statistical Analysis

      Mixed model procedures from SAS 9.1 (
      SAS Institute
      SAS® System Software: Release 9.1.
      ) for a replicated 3 × 3 Latin square design were used to evaluate the effect of treatments on parameters. Raw data were transformed (square root, x2, x3, 1/x, 1/x3) when it was deemed appropriate to meet homogeneity of variance criteria. Cows were defined as a random effect, and Akaike's information criterion was used to select the best covariance structure among compound symmetry, first-order autoregressive, and unstructured. Orthogonal contrasts were defined a priori and used to compare the following: 1) control vs. L-HAY treatments and 2) L-HAY vs. L-HCl treatments. Differences between treatments were declared significant when P-values were < 0.05 and a tendency was noted when 0.05 < P < 0.10.

      Results

      The DCAD of L-HAY was lower than control (P < 0.001) but did not differ from the L-HCl treatment (P = 0.89; Table 2). Compared with control, L-HAY had no effect on BW or BW variation, but average BW was higher with L-HAY than with L-HCl. Dry matter intake and DMI/BW were similar between L-HAY and control, but DMI had a tendency to be lower when cows were fed L-HCl as compared with L-HAY (Table 3). Based on the mean DMI of 9.8 kg/d, cows on L-HCl ingested on average 293 mL of concentrated HCl (12 N) per day. The patterns of DMI during the 14-d experimental periods were similar for cows fed L-HAY and control diets, but the DMI of cows fed L-HCl decreased initially before returning to comparable levels after an average of 7 d (Figure 1). Dry matter of feces and total fecal production were similar between L-HAY and control, whereas DM of feces was higher and total fecal excretion was lower for cows fed L-HCl compared with L-HAY (Table 3). Urine volume of cows did not vary between L-HAY and control nor between L-HCl and L-HAY. Urine pH decreased from 8.21 to 5.89 when L-HAY was fed instead of control (Table 3), but there was no significant difference between L-HCl (5.78) and L-HAY.
      Table 3Body weight, DM intake, feces and urine production, and urine pH of cows fed diets with different levels of DCAD obtained by using a low-DCAD hay or HCl
      Treatment
      Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Contrast
      ItemControlL-HAYL-HClSEP (treatment)Control vs. L-HAYL-HAY vs. L-HCl
      BW and DMI
       BW, kg79379376733.3<0.0010.92<0.001
       BW variation, kg/d1.761.36− 0.370.660.120.740.12
       DMI, kg12.011.59.81.040.070.550.08
       DMI/BW, %1.521.461.270.120.110.560.12
      Feces and urine
       Feces DM, %16.616.517.50.510.060.740.04
       Feces, kg of DM3.383.392.690.258<0.010.96<0.01
       Urine volume, L16.6615.9518.291.660.220.450.10
       Urine pH8.215.895.780.051<0.001<0.0010.13
      1 Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Figure thumbnail gr1
      Figure 1Variation in DMI during the 14 d of the experimental period for cows fed diets with different levels of DCAD obtained by using a low-DCAD hay or HCl. ■ = control; ▴ = L-HAY (low-DCAD treatment using low-DCAD hay); ♦ = L-HCl (low-DCAD treatment using HCl). Error bar = standard error.
      The L-HAY treatment resulted in a tendency for higher blood Ca2+, but there was no difference in blood pH and partial blood pressure of CO2 and O2 when compared with control (Table 4). As well, there was a tendency for lower HCO3, and base excess was lower for cows fed L-HAY compared with control. Blood Cl was higher, and a tendency was observed for higher blood Na+ when L-HAY was fed instead of control. Except for lower blood pH for L-HCl compared with L-HAY, there were no differences in other blood components between cows on these 2 treatments.
      Table 4Blood components in cows fed diets with different levels of DCAD obtained by using a low-DCAD hay or HCl
      Treatment
      Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Contrast
      ItemControlL-HAYL-HClSEP (treatment)Control vs. L-HAYL-HAY vs. L-HCl
      Ca2+, mg/dL5.075.335.330.1150.020.081.00
      pH7.407.377.310.0160.010.160.05
      pCO2,
      pCO2 = partial pressure of CO2.
      mmHg
      42.941.540.91.650.730.610.80
      pO2,
      pO2 = partial pressure of O2.
      mmHg
      36.334.835.71.600.690.400.62
      HCO3,
      Bicarbonate.
      mM
      27.023.022.01.330.070.070.61
      Base excess, mM2.8−1.9−3.11.330.030.040.52
      Na+, mg/dL60.0760.8160.940.2600.100.090.75
      K+, mg/dL0.921.021.020.0310.120.070.91
      Cl, mg/dL29.4830.5230.580.2290.020.020.82
      Anion gap,
      Anion gap = K+ + Na+-Cl−-HCO3−.
      mEq/L
      11.812.613.30.560.200.300.40
      1 Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      2 pCO2 = partial pressure of CO2.
      3 pO2 = partial pressure of O2.
      4 Bicarbonate.
      5 Anion gap = K+ + Na+ - Cl - HCO3.
      Treatments had no effect on EDTA challenge duration or the volume of solution infused to decrease the initial blood Ca2+ by half (Table 5). Cows fed L-HAY had shorter Ca2+ recovery times by over 50% compared with cows fed the control treatment (339 vs. 708 min) and had similar recovery times compared with L-HCl.
      Table 5Duration of EDTA challenge, volume of EDTA solution infused to decrease blood Ca2+ by half, and recovery time needed after the EDTA infusion for blood Ca2+ to reach its initial concentration in cows fed diets with different levels of DCAD obtained by using a low-DCAD hay or HCl
      Treatment
      Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Contrast
      ItemControlL-HAYL-HClSEP (treatment)Control vs. L-HAYL-HAY vs. L-HCl
      Duration, min1031019912.40.940.840.89
      Volume infused, mL81978077084.80.860.680.91
      Ca2+ recovery time, min70833931158.8<0.01<0.010.73
      1 Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      There was no significant difference in intake, apparent digestion, and fecal excretion of ADF, NDF, and N between cows fed L-HAY and control (Table 6). Similarly, differences were not significant for the apparent retention and urine excretion of N between cows fed L-HAY and control (Table 6). Intake, fecal excretion, and total apparent digestion of ADF, NDF, and N were lower in cows fed L-HCl compared with L-HAY. However, there was no significant difference in the digestion of these nutrients when expressed as a percentage of intake for cows on the latter 2 treatments. Apparent digestibility of total DM either expressed as kilograms per day or as percentage of intake did not vary significantly between L-HAY and control, nor between L-HAY and L-HCl diets (Table 6). As well, less N was apparently retained (g/d) and found in urine for L-HCl compared with L-HAY, but both treatments had the same apparent retention when expressed as a percentage of intake.
      Table 6Fiber and N intake; concentration of fiber and N in feces; apparent digestion of fiber, N, and total DM; and N apparent retention in cows fed diets with different levels of DCAD obtained by using a low-DCAD hay or HCl
      Treatment
      Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Contrast
      ItemControlL-HAYL-HClSEP (treatment)Control vs. L-HAYL-HAY vs. L-HCl
      ADF
       Intake, kg/d3.83.73.10.220.030.950.03
       Feces, kg/d1.11.10.90.090.010.800.01
       Digested, kg/d2.82.92.20.310.120.640.06
       Digested, % intake71.171.770.21.440.520.620.27
      NDF
       Intake, kg/d7.16.95.70.590.020.650.02
       Feces, kg/d1.71.71.40.14<0.010.90<0.01
       Digested, kg/d5.45.24.30.470.030.570.04
       Digested, % intake75.875.275.81.180.640.420.42
      N
       Intake, g/d289.3304.6230.129.590.040.520.02
       Feces, g/d76.777.162.66.280.020.930.02
       Digested, g/d212.6227.8167.423.780.050.460.02
       Digested, % intake73.374.472.61.180.530.470.28
       Urine, g/d18.118.315.51.520.050.810.03
       Retained, g/d194.4208.8152.522.920.070.480.03
       Retained, % intake66.768.366.11.610.560.450.30
      Total DM
       Digested, kg/d8.708.317.130.8480.180.630.19
       Digested, % intake71.871.172.31.150.660.600.38
      1 Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Potassium intake was lower for L-HAY as compared with control but did not differ significantly from the L-HCl treatment (Table 7). Apparently absorbed K was lower when expressed in grams per day, and tended to be lower when expressed as a percentage of intake, for L-HAY compared with control (Table 7). When compared with L-HCl, K apparently absorbed, as a percentage of intake, tended to be higher, and K excreted in feces was lower for L-HAY. Potassium excreted in urine was less when cows were fed L-HAY instead of control, which resulted in similar apparently retained K, expressed in grams per day or as a percentage of intake (Table 7). No significant difference was observed between L-HAY and L-HCl for urinated K and apparently retained K expressed in grams per day or as a percentage of intake.
      Table 7Intake, excretion in feces and urine, apparent absorption, and apparent retention of K, Na, Cl, and S in cows fed diets with different levels of DCAD obtained by using a low-DCAD hay or HCl
      Treatment
      Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Contrast
      ItemControlL-HAYL-HClSEP (treatment)Control vs. L-HAYL-HAY vs. L-HCl
      K
       Intake, g/d282.3226.2233.322.320.02<0.010.67
       Feces, g/d28.227.433.73.320.050.750.03
       Absorbed, g/d255.0199.2200.028.600.010.010.96
       Absorbed, % intake90.287.785.30.05<0.010.060.07
       Urine, g/d211.9151.9167.015.55<0.01<0.0010.21
       Retained, g/d42.346.032.88.800.380.700.20
       Retained, % intake15.119.613.12.780.180.200.08
      Na
       Intake, g/d12.69.910.21.570.320.200.89
       Feces, g/d6.83.11.50.950.020.060.19
       Absorbed, g/d3.34.010.01.870.050.810.05
       Absorbed, % intake43.260.086.07.68<0.010.100.04
       Urine, g/d5.57.86.71.020.690.400.67
       Retained, g/d− 1.1− 1.10.61.400.780.860.64
       Retained, % intake− 6.4− 38.1− 2.517.980.690.710.40
      Cl
       Intake, g/d89.7171.2187.219.67<0.01<0.010.46
       Feces, g/d7.69.75.81.420.090.180.03
       Absorbed, g/d81.7162.0180.019.08<0.01<0.010.41
       Absorbed, % intake91.994.496.90.75<0.010.020.02
       Urine, g/d75.1130.6152.511.99<0.01<0.010.14
       Retained, g/d7.331.128.410.100.130.080.83
       Retained, % intake8.116.712.84.630.360.170.51
      S
       Intake, g/d26.129.322.51.580.040.180.01
       Feces, g/d12.212.811.31.080.060.230.02
       Absorbed, g/d14.017.311.02.510.130.230.05
       Absorbed, % intake54.155.050.42.000.110.640.06
       Urine, g/d12.213.311.21.250.340.400.16
       Retained, g/d2.03.20.31.720.450.600.23
       Retained, % intake7.19.50.26.800.560.790.32
      1 Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Sodium intake was similar between L-HAY and the control treatment (Table 7). The L-HAY treatment had a tendency to increase apparently absorbed Na expressed as a percentage of intake, which resulted in a tendency for less Na excreted in feces (Table 7) as compared with control. Sodium apparently absorbed, expressed in grams per day or as a percentage of intake, was lower for L-HAY than L-HCl. There were only numerical differences for Na apparently retained or urinated when cows were fed L-HAY or the control, as well as for Na intake, Na excreted in feces and urine, and Na apparently retained in cows fed L-HAY or L-HCl (Table 7).
      Because the low-DCAD timothy hay contained more Cl (Table 2), Cl intake (Table 7) was higher when cows were fed L-HAY as compared with the control diet. Although Cl concentration was higher in L-HCl than in the L-HAY diet (Table 2), Cl intake was only numerically higher for L-HCl compared with L-HAY (Table 7), because DMI was lower for that treatment (Table 3). Apparent absorption of Cl, expressed in grams per day or as a percentage of intake, was higher for cows fed L-HAY than control (Table 7). Excretion of Cl in feces was lower for L-HCl than for L-HAY, which resulted in a higher Cl apparent absorption expressed as a percentage of intake, but only a numerical increase was observed in the total amount of Cl apparently absorbed (Table 7). When fed L-HAY, cows had higher Cl excretion in urine than the control but a similar amount of apparently retained Cl, expressed in grams per day or as a percentage of intake, compared with control or L-HCl treatments (Table 7).
      No differences were observed between L-HAY and control for S intake, S excreted in feces and urine, and S absorbed or retained. Intake of S was lower for L-HCl compared with L-HAY, which resulted in lower S excreted in feces and apparently absorbed. However, no significant difference was observed for S urinated or retained between these 2 treatments (Table 7).
      The L-HAY treatment had no effect on Ca intake, Ca excreted in feces, and Ca apparently absorbed compared with control (Table 8). Lower Ca intake and Ca excreted in feces were observed for L-HCl compared with L-HAY, but similar amounts of Ca were apparently absorbed. Cows fed the L-HAY treatment excreted more Ca in urine than when they were on control but a similar amount when they were on L-HCl. Cows excreted more Ca than they absorbed for all treatments, which resulted in a negative value for apparently retained Ca. When they were on L-HAY, cows tended to retain less (lose more) Ca (g/d) than when they were fed the control treatment but retained more (lost less) Ca than when they received the L-HCl diet.
      Table 8Intake, excretion in feces and urine, apparent absorption, and apparent retention of Ca, P, and Mg in cows fed diets with different levels of DCAD obtained by using a low-DCAD hay or HCl
      TreatmentContrast
      ItemControlL-HAYL-HClSEP (treatment)Control vs. L-HAYL-HAY vs. L-HCl
      Ca
       Intake, g/d57.361.433.74.30<0.0010.26<0.001
       Feces, g/d58.256.240.84.670.020.700.02
       Absorbed, g/d0.56.12.34.220.550.300.48
       Absorbed, % of intake− 0.48.5− 0.96.70.470.280.32
       Urine, g/d1.213.412.71.72<0.001<0.0010.68
       Retained, g/d− 1.1− 6.1− 12.13.21<0.010.090.05
       Retained, % of intake− 3.6− 10.7− 31.56.89<0.010.27<0.01
      P
       Intake, g/d39.036.532.03.460.060.340.12
       Feces, g/d34.534.527.62.060.010.98<0.01
       Absorbed, g/d3.52.54.42.750.800.710.52
       Absorbed, % of intake10.85.011.65.120.490.330.29
       Urine, g/d0.50.81.90.530.010.280.03
       Retained, g/d3.30.72.60.930.340.160.52
       Retained, % of intake9.31.54.84.060.280.140.51
      Mg
       Intake, g/d15.320.112.61.15<0.010.01<0.01
       Feces, g/d14.016.310.30.92<0.010.03<0.001
       Absorbed, g/d1.03.02.40.560.110.060.61
       Absorbed, % of intake7.116.418.84.220.090.110.66
       Urine, g/d1.12.21.80.220.02<0.010.37
       Retained, g/d0.11.20.60.780.500.260.52
       Retained, % of intake0.65.13.14.140.650.380.70
      1Control = high-DCAD treatment; L-HAY = low-DCAD treatment using low-DCAD hay; L-HCl = low-DCAD treatment using HCl.
      Phosphorus intake and P excreted in feces and urine, apparently absorbed and apparently retained, were similar for L-HAY and the control treatment. Phosphorus intake was similar between L-HCl and L-HAY. Less P was excreted in feces but more in urine when cows were fed L-HCl as compared with L-HAY (Table 8). The variations in P excretion only resulted in numerical differences in P absorbed and retained.
      Magnesium intake and Mg excreted in feces were higher for L-HAY as compared with both control and L-HCl treatments (Table 8). As well, apparently absorbed Mg, expressed in grams per day, tended to be higher and Mg excreted in urine was higher for L-HAY when compared with control but not with L-HCl. Apparently absorbed Mg, expressed as a percentage of intake, and apparently retained Mg, expressed in grams per day or as a percentage of intake, did not vary when cows were fed L-HAY instead of control or L-HCl treatments.

      Discussion

      Effect of Low-DCAD Hay to Decrease the Ration DCAD

      Comparison between L-HAY and control treatment indicates that low-DCAD hay can induce a compensated metabolic acidosis as shown by decreases in urinary pH and blood HCO3 and base excess. This acidification capacity resulted in a tendency for increased blood Ca2+ and a greater ability to increase Ca2+ availability when cows are stressed, like during the EDTA challenge. Differences in the absorption of some minerals were observed when cows were fed L-HAY instead of the control diet, but there was no difference in DMI and digestion of OM, ADF, NDF, and N, which should not be altered during the critical period of transition. These results support the interest of using a low-DCAD hay before calving.
      Because urinary pH is directly related to DCAD (
      • Vagnoni D.B.
      • Oetzel G.R.
      Effects of dietary cation-anion difference on the acid-base status of dry cows.
      ;
      • Charbonneau E.
      • Pellerin D.
      • Oetzel G.R.
      Impact of lowering dietary cation-anion difference in nonlactating dairy cows: A meta-analysis.
      ), it was expected that the low-DCAD hay would diminish urinary pH. Lowering the DCAD of a ration to - 20 mEq/kg is within the recommended levels of 0 mEq/kg or lower (
      • Horst R.L.
      • Goff J.P.
      Milk fever and dietary potassium.
      ;
      NRC
      Nutrient Requirements of Dairy Cattle.
      ;
      • Charbonneau E.
      • Pellerin D.
      • Oetzel G.R.
      Impact of lowering dietary cation-anion difference in nonlactating dairy cows: A meta-analysis.
      ), but this resulted in a urinary pH that was lower than the recommended level of 7.0 (
      NRC
      Nutrient Requirements of Dairy Cattle.
      ;
      • Charbonneau E.
      • Pellerin D.
      • Oetzel G.R.
      Impact of lowering dietary cation-anion difference in nonlactating dairy cows: A meta-analysis.
      ). Low urinary pH is associated with too strong a metabolic acidosis for the cow (
      • Charbonneau E.
      • Pellerin D.
      • Oetzel G.R.
      Impact of lowering dietary cation-anion difference in nonlactating dairy cows: A meta-analysis.
      ). The great difference we observed in urinary pH in response to the DCAD value of the ration in this trial emphasizes the importance of testing urinary pH to evaluate the reaction of each cow to low-DCAD rations.
      Lower blood HCO3 and base excess observed with the L-HAY treatment indicate a compensated metabolic acidosis resulting in a higher level of blood Ca2+ (
      • Joyce P.W.
      • Sanchez W.K.
      • Goff J.P.
      Effect of anionic salts in prepartum diets based on alfalfa.
      ;
      • Rose D.B.
      • Post T.W.
      Clinical Physiology of Acid-Base and Electrolyte Disorders.
      ;
      • Goff J.P.
      • Ruiz R.
      • Horst R.L.
      Relative acidifying activity of anionic salts commonly used to prevent milk fever.
      ), even without Ca deficit. The lack of measurable difference for partial blood pressure of CO2 and O2 indicates that the acidosis did not result in respiratory mechanisms to compensate for the metabolic acidosis (
      • Rose D.B.
      • Post T.W.
      Clinical Physiology of Acid-Base and Electrolyte Disorders.
      ). Higher blood Cl probably comes from the increase in Cl apparently absorbed, whereas the tendency for higher blood Na+ might come from the close regulation of Na with Cl (
      • Shills M.E.
      Magnesium.
      ).
      Because blood Ca2+ tended to be higher when cows were fed L-HAY compared with control, cows on this treatment might have shown more resistance to hypocalcemia than those fed the control diet, as demonstrated by
      • Wang C.
      • Beede D.K.
      Effects of supplemental protein on acid-base status and calcium metabolism of nonlactating Jersey cows.
      with an EDTA infusion. In our EDTA infusion challenge, however, this increase in resistance was not confirmed. The difference in response between both experiments could be due to different protocols.
      • Wang C.
      • Beede D.K.
      Effects of supplemental protein on acid-base status and calcium metabolism of nonlactating Jersey cows.
      stopped their EDTA infusion when 1 of the cows in the group showed signs of milk fever, whereas we stopped our infusion when individual cows reached 50% of their initial blood Ca2+ level. As well, the lack of response for time of infusion and volume infused, as opposed to the results of
      • Schonewille J.T.
      • Van’t Klooster A.T.
      • Wouterse H.
      • Beynen A.C.
      Hypocalcemia induced by intravenous administration of disodium ethylenediaminotetraacetate and its effects on excretion of calcium in urine of cows fed a high chloride diet.
      , could probably be explained by the EDTA solution being more concentrated in our experiment (7 vs. 5%) with a similar rate of infusion. In our experiment, all cows responded similarly to the infusion. The lower concentration of EDTA solution used in
      • Schonewille J.T.
      • Van’t Klooster A.T.
      • Wouterse H.
      • Beynen A.C.
      Hypocalcemia induced by intravenous administration of disodium ethylenediaminotetraacetate and its effects on excretion of calcium in urine of cows fed a high chloride diet.
      may have allowed a wider range of responses from the different treatments. The decrease in blood Ca2+ recovery time after the EDTA challenge observed in cows fed L-HAY or L-HCl is usual for low-DCAD treatments (
      • Schonewille J.T.
      • Van’t Klooster A.T.
      • Wouterse H.
      • Beynen A.C.
      Hypocalcemia induced by intravenous administration of disodium ethylenediaminotetraacetate and its effects on excretion of calcium in urine of cows fed a high chloride diet.
      ) and confirms that a low-DCAD timothy hay is effective in increasing Ca2+ availability when cows are in need.
      An increase in Ca absorption is usually associated with Ca regulation when cows are in hypocalcemia (
      • Horst R.L.
      Regulation of calcium and phosphorus homeostasis in the dairy cow.
      ). In the present experiment, nonpregnant dry cows were used, and Ca requirements were far lower than those required for colostrum production at calving; this could explain the lack of difference in apparent absorption of Ca. As well, the low apparent absorption of Ca (negative for 2 of the 3 treatments) and Ca retention as a percentage of intake (negative for all treatments) were similar to those observed by
      • Wang C.
      • Beede D.K.
      Effects of supplemental protein on acid-base status and calcium metabolism of nonlactating Jersey cows.
      and
      • Schonewille J.T.
      • Van’t Klooster A.T.
      • Beynen A.C.
      The addition of extra calcium to a chloride-rich ration does not affect the absolute amount of calcium absorbed by non-pregnant, dry cows.
      and would mostly come from endogenous Ca excreted into the gut lumen (
      • Moodie E.W.
      Some aspects of hypocalcemia in cattle.
      ;
      • Wang C.
      • Beede D.K.
      Effects of supplemental protein on acid-base status and calcium metabolism of nonlactating Jersey cows.
      ). Endogenous excretions have more effect on multiparous cows, because they excrete as much as younger animals but do not absorb Ca as efficiently (
      • Moodie E.W.
      Some aspects of hypocalcemia in cattle.
      ;
      • Wang C.
      • Beede D.K.
      Effects of supplemental protein on acid-base status and calcium metabolism of nonlactating Jersey cows.
      ). An increase in Ca excretion in urine is typical of a low-DCAD ration (
      • Gaynor P.J.
      • Mueller F.J.
      • Miller J.K.
      • Ramsey N.
      • Goff J.P.
      • Horst R.L.
      Parturient hypocalcemia in Jersey cows fed alfalfa haylage-based diets with different cation to anion ratios.
      ;
      • Van Mosel M.
      • Van’t Klooster A.T.
      • Van Mosel F.
      • Van der Kuilen J.
      Effects of reducing dietary [(Na+ + K+) (Cl + SO4] on the rate of calcium mobilisation by dairy cows at parturition.
      ;
      • Schonewille J.T.
      • Van’t Klooster A.T.
      • Beynen A.C.
      The addition of extra calcium to a chloride-rich ration does not affect the absolute amount of calcium absorbed by non-pregnant, dry cows.
      ). The increase of Ca in urine (Table 8) and the tendency for higher blood Ca (Table 5) with no increase in Ca apparent absorption indicates that Ca mobilization from reserves is responsible for the increase in available Ca with L-HAY compared with control. The increase in Ca available without absorption difference supports the hypothesis that cows, even without Ca deficit, are mobilizing Ca from bone when fed a low-DCAD ration.
      The lower K intake for L-HAY could explain the lower K apparent absorption, because K is mostly absorbed by diffusion (
      NRC
      Nutrient Requirements of Dairy Cattle.
      ). The similar apparently retained K but lower K intake for the L-HAY treatment compared with the control confirms that K is regulated mainly through urination.
      Both L-HAY and control treatments resulted in a low Na apparent absorption, expressed as a percentage of intake. Sodium is considered to be easily absorbed in the digestive tract with absorption at around 90% (
      NRC
      Nutrient Requirements of Dairy Cattle.
      ); it is regulated mostly through renal mechanisms. In this experiment, cows fed the control diet had low Na apparent absorption (43% of intake). The Na apparent absorption was better for cows fed the L-HAY diet (60%), but this level is still far below the expected absorption of 90%, especially given that the Na of both diets was in large part from salt (NaCl), which is considered to be absorbed at nearly 100% (
      NRC
      Nutrient Requirements of Dairy Cattle.
      ). Studies specifically on dry cows have reported variations in apparent absorption of Na between 55 and 83% of intake (
      • Leclerc H.
      • Block E.
      Effects of reducing dietary cation-anion balance for prepartum dairy cows with specific reference to hypocalcemic parturient paresis.
      ;
      • Delaquis A.M.
      • Block E.
      Acid-base status, renal function, water, and macromineral metabolism of dry cows fed diets differing in cation-anion difference.
      ). Results from the current study and previous ones suggest that Na apparent absorption is lower for nonlactating dairy cows as compared with lactating cows. Because Cl regulation is mainly through urine, and Cl is mostly absorbed in the gut (
      • Underwood E.J.
      • Suttle N.F.
      The Mineral Nutrition of Livestock.
      ), the difference in Cl apparent absorption would typically come from differences in intake.
      Potassium is known to affect Mg absorption (
      • Schonewille J.T.
      • Van’t Klooster A.T.
      • Wouterse H.
      • Beynen A.C.
      Effects of intrinsic potassium in artificially dried grass and supplemental potassium bicarbonate on apparent magnesium absorption in dry cows.
      ), but the variation in K concentration between diets was not sufficient to have had an effect. The variation in K concentration between both forages was smaller than that observed by
      • Schonewille J.T.
      • Van’t Klooster A.T.
      • Wouterse H.
      • Beynen A.C.
      Effects of intrinsic potassium in artificially dried grass and supplemental potassium bicarbonate on apparent magnesium absorption in dry cows.
      , who reported a decrease of Mg absorption when forage with high K (4.75% of DM) was fed instead of forage with low K (2.75% of DM). The difference of Mg concentration in feces and urine could be due to increased Mg intake, which resulted in more Mg excretion from intestinal and renal regulation. Magnesium homeostasis is known to be affected by both intestinal and renal regulation (
      • Shills M.E.
      Magnesium.
      ). The urine excretion results are in accordance with
      • Wang C.
      • Beede D.K.
      Effects of supplemental protein on acid-base status and calcium metabolism of nonlactating Jersey cows.
      , who reported an increase in urine Mg excretion associated with an increase in protein and Mg concentration in the diet. Renal regulation of Mg is not fully understood and could also be influenced by NaCl reabsorption (
      • Rose D.B.
      • Post T.W.
      Clinical Physiology of Acid-Base and Electrolyte Disorders.
      ); this may also partly explain the increased Mg excretion with L-HAY compared with control, because L-HAY also increased Cl excretion.

      Comparison of Low-DCAD Treatments

      Comparison between L-HAY and L-HCl shows that decreasing DCAD with a low-DCAD hay can be as effective as decreasing it with HCl. There was little difference between the 2 treatments for urinary pH, blood components, and in the EDTA challenge. Considering the length of time it took for cows to adapt to the L-HCl diet, and the tendency of reduced DMI to result in lower absorption of ADF, NDF, and N when L-HCl was fed instead of L-HAY, low-DCAD hay might be a better choice than dietary HCl addition for transition cows given their high nutritional needs. The DCAD of L-HCl ration had already been lowered by the addition of 23% of low-DCAD hay. Even with the mixture of forages used, the amount of HCl (±293 mL, which represent 2.5 Eq/d) required to achieve similar DCAD then L-HAY was apparently enough to decrease DMI. The tendency for lower DMI for L-HCl compared with L-HAY confirms that a small augmentation over the 2.3 Eq/d identified from
      • Oetzel G.R.
      • Barmore J.A.
      Intake of a concentrate mixture containing various anionic salts fed to pregnant, nonlactating dairy cows.
      can effectively result in lower DMI. These results suggest that using low-DCAD hay instead of HCl to decrease DCAD of the ration by the same magnitude does not have the same effect in DMI.
      The difference in DMI could explain the variations in BW observed. Two-week periods are short to determine BW variations, but results were still interpreted, because the L-HCl treatment had a very strong effect on DMI, which resulted in differences in BW. The higher DM concentration in feces with the L-HCl ration suggests there is an increase in water absorption when HCl is used to decrease the DCAD value of a ration.
      The lack of difference between L-HAY and L-HCl treatments during the EDTA challenge corroborates the hypothesis that low-DCAD hay can be as efficient as HCl at decreasing the risk of hypocalcemia. Because blood pH was less affected by L-HAY than L-HCl, low-DCAD hay might work as well as HCl at decreasing the risk of hypocalcemia while also not being as stressful on cow metabolism.
      A higher intake of Ca for the L-HAY treatment compared with the L-HCl treatment is associated with a higher DMI for L-HAY and a higher Ca concentration in the low-DCAD as compared with the high-DCAD timothy hay. The difference in Ca intake but not in Ca absorption resulted in more Ca excreted in feces for L-HAY as compared with L-HCl. Cows on both treatments excreted more Ca in urine than their Ca intakes, which is typical of low-DCAD diets (
      • Gaynor P.J.
      • Mueller F.J.
      • Miller J.K.
      • Ramsey N.
      • Goff J.P.
      • Horst R.L.
      Parturient hypocalcemia in Jersey cows fed alfalfa haylage-based diets with different cation to anion ratios.
      ;
      • Van Mosel M.
      • Van’t Klooster A.T.
      • Van Mosel F.
      • Van der Kuilen J.
      Effects of reducing dietary [(Na+ + K+) (Cl + SO4] on the rate of calcium mobilisation by dairy cows at parturition.
      ;
      • Schonewille J.T.
      • Van’t Klooster A.T.
      • Beynen A.C.
      The addition of extra calcium to a chloride-rich ration does not affect the absolute amount of calcium absorbed by non-pregnant, dry cows.
      ). More Ca was apparently retained with L-HAY compared with L-HCl. The higher Ca retained for L-HAY probably comes from the numerically higher absorbed Ca (+3.8 g/d) with L-HAY than L-HCl but a similar excretion of Ca in urine (+0.7 g/d). The lower Ca retention for L-HAY could indicate a lower Ca mobilization as compared with L-HCl, which would be consistent with the higher blood pH for L-HAY but not with the identical level of blood Ca2+.
      Although there were higher concentrations of K in the L-HCl diet, the tendency for lower DMI for cows on this treatment resulted in no difference in K intake compared with L-HAY. Less K was excreted in feces for L-HAY as compared with L-HCl, and when linked with the numerically lower K intake, this explains the tendency for lower apparently absorbed K expressed as a percentage of intake but the lack of difference in apparently absorbed K expressed in grams per day.
      When considering the processes of Na and Cl absorption, the difference between L-HAY and L-HCl in Na total apparent absorption (g/d) and Na and Cl apparent absorption as a percentage of intake could be explained, in part, by an increase in the absorption of these 2 minerals in the form of NaCl (
      • Harper M.E.
      • Willis J.S.
      • Patrick J.
      Sodium and chloride in nutrition.
      ). A complementary mechanism to NaCl absorption for Na and Cl exists in the exchange of Na+ for H+ and Cl for HCO3 (
      • Harper M.E.
      • Willis J.S.
      • Patrick J.
      Sodium and chloride in nutrition.
      ). Because H+ and HCO3 come from the chemical reaction of transforming the CO2 entering the cell [CO2 + H2O ⇒H+ + HCO3], the exchange of H+ and HCO3 in the intestinal cells for Na+ and Cl from the gut lumen provides neutrality. Because both diets were high in Cl, more Cl than Na would have been exchanged by this process, which must have disturbed the equilibrium between H+ and HCO3. The absorption mechanism in the gut lumen and the mechanism in blood to maintain electrical neutrality could explain the metabolic acidosis created by both treatments.

      Conclusions

      Results from this study indicate that low-DCAD hay effectively reduces the DCAD value of dry cow diets and modifies cow metabolism as indicated by reduced urine pH, difference in blood components, and a better response to simulation of hypocalcemia when compared with diets without DCAD alteration. When compared with HCl, a product used to decrease the DCAD of a ration, the low-DCAD hay obtained similar results without the adverse effect of HCl addition on DMI. A low-DCAD hay could provide a viable alternative to preventing hypocalcemia at calving.

      Acknowledgments

      This study was supported by a grant from the Action Concertée FQRNT-Novalait-MAPAQ in collaboration with Agriculture and Agri-Food Canada and a scholarship from FQRNT. We thank Annie Brégard for her help during the experiment. Also, sincere appreciation is expressed to the staff of the Centre de recherche en sciences animales de Deschambault, the institution where the experiment took place. Thanks go to Mario Laterrière from Agriculture and Agri-Food Canada, Québec, for laboratory assistance. Appreciation is also extended to Rachel Gervais and Andrée-Anne Gingras, graduate students, as well as François Bécotte, undergraduate student, at Université Laval for their help during sample collections.

      References

        • Beede D.K.
        • Risco C.A.
        • Donovan G.A.
        • Wang C.
        • Archbald L.F.
        • Sanchez W.K.
        Nutritional management of the late pregnant dry cow with particular reference to dietary cationanion difference and calcium supplementation.
        Proc. 24th Annu. Conf. Am. Assoc. Bovine Pract., Stillwater, OK1992 (Pages 51–55)
        • Beede D.K.
        • Sanchez W.K.
        • Wang C.
        Macrominerals.
        in: Van Horn H.H. Wilcox C.J. Large Dairy Herd Management. Univ. Florida, Gainesville1992: 272-286
        • Block E.
        Manipulating dietary anions and cations for prepartum dairy cows to reduce incidence of milk fever.
        J. Dairy Sci. 1984; 67: 2939-2948
        • Canadian Council of Animal Care
        2nd ed. Guide to the Care and Use of Experimental Animals. 1. Can. Counc. Anim. Care, Ottawa, Ontario, Canada1993
        • Charbonneau E.
        • Pellerin D.
        • Oetzel G.R.
        Impact of lowering dietary cation-anion difference in nonlactating dairy cows: A meta-analysis.
        J. Dairy Sci. 2006; 89: 537-548
        • Curtis C.R.
        • Erb H.N.
        • Sniffen C.J.
        • Smith R.D.
        • Powers P.A.
        • Smith M.C.
        • White M.E.
        • Hillman R.B.
        • Pearson E.J.
        Association of parturient hypocalcemia with eight periarturient disorders in Holstein cows.
        JAVMA. 1983; 5: 559-561
        • Delaquis A.M.
        • Block E.
        Acid-base status, renal function, water, and macromineral metabolism of dry cows fed diets differing in cation-anion difference.
        J. Dairy Sci. 1995; 78: 604-619
        • Dishington I.W.
        Prevention of milk fever (hypocalcemic paresis puerperalis) by dietary salt supplements.
        Acta Vet. Scand. 1975; 16: 503-512
        • Gaynor P.J.
        • Mueller F.J.
        • Miller J.K.
        • Ramsey N.
        • Goff J.P.
        • Horst R.L.
        Parturient hypocalcemia in Jersey cows fed alfalfa haylage-based diets with different cation to anion ratios.
        J. Dairy Sci. 1989; 72: 2525-2531
        • Goff J.P.
        • Horst R.L.
        Use of hydrochloric acid as a source of anions for prevention of milk fever.
        J. Dairy Sci. 1998; 81: 2874-2880
        • Goff J.P.
        • Horst R.L.
        • Jardon P.W.
        • Borelli C.
        • Wedam J.
        Field trials of an oral calcium propionate paste as an aid to prevent milk fever in periparturient dairy cows.
        J. Dairy Sci. 1996; 79: 378-383
        • Goff J.P.
        • Ruiz R.
        • Horst R.L.
        Relative acidifying activity of anionic salts commonly used to prevent milk fever.
        J. Dairy Sci. 2004; 87: 1245-1255
        • Gröhn Y.T.
        • Erb H.N.
        • McCulloch C.E.
        • Saloniemi H.S.
        Epidemiology of metabolic disorders in dairy cattle: Association among host characteristics, disease, and production.
        J. Dairy Sci. 1989; 72: 1876-1885
        • Harper M.E.
        • Willis J.S.
        • Patrick J.
        Sodium and chloride in nutrition.
        in: O’Dell B.L. Sunde R.A. Handbook of Nutritionally Essential Mineral Elements. Marcel Dekker Inc., New York, NY1997: 93-116
        • Horst R.L.
        Regulation of calcium and phosphorus homeostasis in the dairy cow.
        J. Dairy Sci. 1986; 69: 604-616
        • Horst R.L.
        • Goff J.P.
        Milk fever and dietary potassium.
        Proc. 1997 Cornell Nutr. Conf. Feed Manuf. Rochester, NY. Cornell Univ., Ithaca, NY1997 (Pages 181–189)
        • Horst R.L.
        • Goff J.P.
        • Reinhardt T.A.
        • Buxton D.R.
        Strategies for preventing milk fever in dairy cattle.
        J. Dairy Sci. 1997; 80: 1269-1280
        • Huber T.L.
        • Wilson R.C.
        • Stattelman A.J.
        • Goetsch D.D.
        Effect of hypocalcemia on motility of the ruminant stomach.
        Am. J. Vet. Res. 1981; 42: 1488-1489
        • Isaac R.A.
        • Johnson W.C.
        Determination of total nitrogen in plant tissue, using a block digestor.
        J. Assoc. Off. Anal. Chem. 1976; 59: 98-100
        • Joyce P.W.
        • Sanchez W.K.
        • Goff J.P.
        Effect of anionic salts in prepartum diets based on alfalfa.
        J. Dairy Sci. 1997; 80: 2866-2875
        • Kume S.
        • Toharmat T.
        • Nonaka K.
        • Oshita T.
        • Nakui T.
        • Ternouth J.H.
        Relationship between crude protein and mineral concentrations in alfalfa and value of alfalfa silage as a mineral source for periparturient cows.
        Anim. Feed Sci. Technol. 2001; 93: 157-168
      1. Lachat Instruments. 2005; Subject: Method list for automated ion analyzers (flow injection analyses, ion chromatography). http://www.lachatinstruments.com/applications/MethodsList.PDF Accessed April 18, 2005.

        • Leclerc H.
        • Block E.
        Effects of reducing dietary cation-anion balance for prepartum dairy cows with specific reference to hypocalcemic parturient paresis.
        Can. J. Anim. Sci. 1989; 69: 411-423
        • Liu L.
        Determination of chloride in plant tissue.
        in: Karla Y.P. Handbook of Reference Methods for Plant Analysis. CRC Press, Boca Raton, FL1998 (Pages 111–113)
        • Miller R.O.
        High temperature oxidation.
        in: Kalra Y.P. Handbook of Reference Methods for Plant Analysis. CRC Press, Boca Raton, FL1998 (Pages 53–56)
        • Mills H.A.
        • Jones Jr., J.B.
        Plant Analysis Handbook II: A Practical Sampling, Preparation, Analysis, and Interpretation Guide.
        Rev. ed. MicroMacro Publ. Inc., Athens, GA1996
        • Moodie E.W.
        Some aspects of hypocalcemia in cattle.
        Vet. Rec. 1960; 72: 1145-1149
        • NRC
        Nutrient Requirements of Dairy Cattle.
        7th ed. Natl. Acad. Press, Washington, DC2001
        • Oetzel G.R.
        • Barmore J.A.
        Intake of a concentrate mixture containing various anionic salts fed to pregnant, nonlactating dairy cows.
        J. Dairy Sci. 1993; 76: 1617-1623
        • Oetzel G.R.
        • Olson J.D.
        • Curtis C.R.
        • Fettman M.J.
        Ammonium chloride and ammonium sulfate for prevention of parturient paresis in dairy cows.
        J. Dairy Sci. 1988; 71: 3302-3309
        • Pelletier S.
        • Bélanger G.
        • Tremblay G.F.
        • Séguin P.
        • Drapeau R.
        • Allard G.
        Dietary cation-anion difference of Timothy (Pleum pratense L.) as influenced by application of chloride and nitrogen fertilizer.
        Grass Forage Sci. 2007; 62: 66-77
        • Rose D.B.
        • Post T.W.
        Clinical Physiology of Acid-Base and Electrolyte Disorders.
        5th ed. McGraw-Hill, New York, NY2001
        • SAS Institute
        SAS® System Software: Release 9.1.
        SAS Inst. Inc., Cary, NC2002 (TS1M3)
        • Schonewille J.T.
        • Van’t Klooster A.T.
        • Beynen A.C.
        The addition of extra calcium to a chloride-rich ration does not affect the absolute amount of calcium absorbed by non-pregnant, dry cows.
        J. Anim. Physiol. Anim. Nutr. 1994; 72: 272-280
        • Schonewille J.T.
        • Van’t Klooster A.T.
        • Dirkzwager A.
        • Beynen A.C.
        Stimulatory effect of an anion (chloride)-rich ration on apparent calcium absorption in dairy cows.
        Livest. Prod. Sci. 1994; 40: 233-240
        • Schonewille J.T.
        • Van’t Klooster A.T.
        • Wouterse H.
        • Beynen A.C.
        Effects of intrinsic potassium in artificially dried grass and supplemental potassium bicarbonate on apparent magnesium absorption in dry cows.
        J. Dairy Sci. 1999; 82: 1824-1830
        • Schonewille J.T.
        • Van’t Klooster A.T.
        • Wouterse H.
        • Beynen A.C.
        Hypocalcemia induced by intravenous administration of disodium ethylenediaminotetraacetate and its effects on excretion of calcium in urine of cows fed a high chloride diet.
        J. Dairy Sci. 1999; 82: 1317-1324
        • Shills M.E.
        Magnesium.
        in: O’Dell B.L. Sunde R.A. Handbook of Nutritionally Essential Mineral Elements. Marcel Dekker Inc., New York, NY1997: 93-116
        • Tremblay G.F.
        • Brassard H.
        • Bélanger G.
        • Séguin P.
        • Drapeau R.
        • Bregard A.
        • Michaud R.
        • Allard G.
        Dietary cation anion difference of five cool-season grasses.
        Agron. J. 2006; 98: 339-348
        • Underwood E.J.
        • Suttle N.F.
        The Mineral Nutrition of Livestock.
        3rd ed. CABI Publ., New York, NY1999
        • USDA
        Part 1: Reference of dairy health and management in the United States, 2002.
        Cent. Epidemiol. Anim. Health, Fort Collins, Co., 2002
        • Vagnoni D.B.
        • Oetzel G.R.
        Effects of dietary cation-anion difference on the acid-base status of dry cows.
        J. Dairy Sci. 1998; 81: 1643-1652
        • Van Mosel M.
        • Van’t Klooster A.T.
        • Van Mosel F.
        • Van der Kuilen J.
        Effects of reducing dietary [(Na+ + K+) (Cl + SO4] on the rate of calcium mobilisation by dairy cows at parturition.
        Res. Vet. Sci. 1993; 54: 1-9
        • Van Soest P.J.
        • Robertson J.B.
        • Lewis B.A.
        Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition.
        J. Dairy Sci. 1991; 74: 3583-3597
        • Wang C.
        • Beede D.K.
        Effects of supplemental protein on acid-base status and calcium metabolism of nonlactating Jersey cows.
        J. Dairy Sci. 1990; 73: 3178-3186