| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL PRODUCTION |
,2* Prairie Swine Centre Inc., Saskatoon, Saskatchewan S7H 5N9, Canada; and and Department of Animal and Poultry Science, University of Saskatchewan, Saskatoon, Saskatchewan S7N 5B5, Canada
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
Key Words: mineral • protein • swine • water
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
). Increasingly, society is viewing water as a limited resource and the livestock industry as a major consumer. To the pork producer, excessive water utilization produces excess manure that is costly to store and to apply on the land. As a result, there is increasing interest in ways to reduce water utilization without adversely affecting the pig; diet modification is one possible strategy.
Various environmental (Patience et al., 2005) and physiological factors (Mroz et al., 1995) affect water utilization; the impact of diet is not clear. Suzuki et al. (1998) and Pfeiffer et al. (1995) showed that water intake increased in response to increasing dietary CP, whereas Albar and Granier (1996) found that barrows, but not gilts, offered a low CP diet had reduced water intake in one study, but differences were not significant in another. These inconsistent results justify further investigation of the relationship between diet composition and water utilization.
Excessive protein intake increases daily urine volume, possibly due to the need to remove excess N from the body; excessive minerals should have the same effect. This assumes a relatively constant urine osmolality across diverse nutrient intakes, an assumption that may not be correct (Pond and Houpt, 1978). If the acceptable range of urine osmolality is wide, the need to adjust urine output and thus water consumption is reduced, and diet will have little impact on water consumption.
We hypothesized that the volume of water consumed and excreted would increase as dietary CP or mineral concentration increases. The objective of this experiment was to determine if increasing dietary CP or mineral would increase water intake and excretion and, conversely, if reducing dietary CP would have the opposite effect.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
. Pigs employed in this experiment were of a similar genotype (C22 female x L-337 male, PIC Canada Ltd., Airdrie, Alberta, Canada), and had an average initial age of 70 ± 2 d and an average initial BW of 34.3 ± 4.6 kg (means ± SD). Before the experiment, the pigs received typical commercial feeds during the nursery and early grower periods.
Drinking water was obtained from the City of Saskatoon supply, which itself is derived from the South Saskatchewan River. Assays of this water, conducted at an approved commercial laboratory (Enviro-Test Laboratories, Saskatoon, Saskatchewan, Canada), revealed low levels of all nutrients, including total dissolved solids (218 mg/L), Na (24 mg/L), Mg (16 mg/L), K (3 mg/L), Cl (9 mg/L), and sulphate (60 mg/L). Nitrate levels were below detectable limits, hardness was 129 mg/L (reported as CaCO3 equivalents), and the pH was 8.2. This water would be considered to be of high quality, and the small concentrations of solutes would not be expected to confound the outcome of this experiment (Patience, 1997).
Facilities
Barrows were housed in individual metabolism pens measuring 1.5 x 1.5 m, allowing complete freedom of animal movement, even during the fecal and urine collection periods. Each metabolism pen was fitted with a single nipple/bowl combination drinker (Drink-O-Mat, Prairie Pride Enterprises, Winnipeg, Manitoba, Canada) to allow continuous access to water. The combination of spillage and evaporative losses with this system were determined and found to be less than 1% of total intake, and therefore are not included in the subsequent calculations.
The ambient temperature and humidity in the experimental rooms were continuously monitored. Conducted during the autumn months, the ambient room temperature was 17 to 18°C. Relative humidity measurements were conducted daily near the center of the room using a psychrometer (Cole-Palmer, Anjou, Quebec, Canada) and ranged from 46 to 58%. Lights were on for 12 h each day.
Experimental Design
Within each of the 3 replicates, 4 pigs per treatment, or 16 in total, were selected for the experiment on the basis of similar age and BW. Three replicates provided a total of 12 observations/treatment. The experimental period within each replicate was 14 d; although this was longer than typical pig metabolism experiments, it allowed the pigs extra time to adjust to the experimental treatments. To reduce the possible impact of boredom on ad libitum water intake, each pen was enriched with a length of chain and a moveable object such as a ball or rubber boot, which were changed twice daily. Pigs were weighed individually on d 1 and 14.
There were 4 dietary treatments (Table 1). Each contained (as-fed basis) 3.3 Mcal of DE/kg and 2.4 g of apparent digestible lysine/Mcal of DE, except for the excess protein diet, which contained 3.4 Mcal of DE/kg and 3.4 g of apparent digestible lysine/Mcal of DE. In all cases, the level of lysine and other AA was sufficient to support the level of protein deposition observed in this experiment. All diets were formulated to meet or exceed the pigs’ nutrient requirements, as defined by the NRC (1998).
|
). The excess protein (ExcP) diet contained sufficient additional AA to support approximately 50% greater growth than that of the HiP and LoP diets; it contained 25.4% CP. The fourth diet (ExcM) contained excessive Ca (1.00 vs. 0.51%), P (0.76 vs. 0.51%), Na (0.36 vs. 0.14%), and K (0.70 vs. 0.28%), compared with the HiP diet. All diets were provided to the pigs in pelleted form.
Sample Collection
Voided urine was captured through a tray and funnel system located directly under each pen. Plastic-mesh screens were installed over the urine tray, and glass wool was inserted in the neck of the funnel to facilitate collection of urine with minimal fecal contamination. Total urine output from each pig was collected into a plastic 4-L container, in which 20 mL of 12 N HCl was added to minimize the loss of ammonium N. The total quantity of voided urine was harvested twice daily, and 1% aliquots frozen at –20°C and retained for later analysis. Total collection of urine and feces occurred on the final 3 d of each dietary period (d 12 to 14, inclusive).
Feces were collected into a plastic bag attached to the pig using a Velcro support system, as described by van Kleef et al. (1994). Feces were harvested and weighed at least twice daily and frozen at –20°C. Feces from each pig were pooled and were homogenized at the end of the collection period using a small vertical mixer (Hobart Corp., Troy, OH), subsampled, and stored at –20°C for later analysis.
Water Balance
Water intake was recorded using individual water meters (Model C7001, ABB Water Meters, Ocala, FL) installed on each metabolism pen. The meters had an accuracy of ± 1.5%; according to the manufacturer, the accuracy of the water meters may be affected by flow rates less than 500 mL/min, and so the nipples were adjusted to ensure a flow rate of 2 to 3 L/min. As a further test of the accuracy of the measurement of water usage, a single meter was installed at the front of the room, through which the total amount of water to all of the individual meters was measured. Water disappearance, as measured by each individual meter, was totaled, and the difference between the sum of all of the individual water meters and that recorded from the main water meter was less than 3%. Water in the feed was determined from its DM content. Fecal water output was determined directly from fecal DM.
Sample Assays
Fecal samples were freeze-dried in a model 40, Virtis freeze-dryer (Virtis Co. Ltd., Gardiner, NY). Samples of diet and freeze-dried feces were ground through a 1.0-mm screen in a Retsch mill (Model ZM1, Brinkman Instruments, Westbury, NY). Ground feces and feed samples were analyzed for DM content in an airflow-type drying oven at 135°C for 2 h (method 930.15; AOAC, 1990). Feed and fecal samples were analyzed for Ca, P, K, and Na using atomic absorption spectroscopy (method 968.08; AOAC, 1990). Feed, fecal, and urine samples were analyzed for N content by the combustion method (method 968.06 AOAC, 1990), using a Leco N determinator (Model FP-528, Leco Co., St. Joseph, MI).
Phosphorus content of urine samples was determined spectrophotometrically at 405 nm using a modification of the method of Newkirk and Classen (1998). Urine samples were analyzed for Cl concentration by indirect titration according to LaCroix et al. (1970), for Ca content by atomic absorption spectroscopy according to Fernandez and Kahn (1971), and for Na and K content by flame photometry (Instrumentation Laboratory, Lexington, MA), using LiOH to suppress Na ionization.
Statistical Analysis
Data were analyzed by ANOVA using the GLM procedure (SAS Inst. Inc., Cary, NC), appropriate to the randomized complete block design. The model included effects of diet, replicate, and a diet x replicate interaction. When the F-test was significant (P < 0.05), means comparisons were performed using the PDIFF option of SAS. Data are reported as least squares means.
The relationship between average daily water intake (ADWI) and ADFI, and between ADWI and daily intake (g/d) or dietary concentration (%) of N, Ca, P, Na, Cl, and K were determined using PROC CORR of SAS. Stepwise regression was applied where appropriate.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
Water Balance
Average daily water intake and urine output tended (P < 0.10) to increase in pigs given the ExcP diet relative to those receiving the HiP diet (Table 2). Furthermore, the water:feed ratio was elevated (P < 0.05) in pigs fed the diet containing excess CP. Fecal water excretion was elevated (P < 0.05) in pigs given the ExcM diet as compared with pigs fed the LoP and HiP diets.
|
). There were no other statistically significant relationships between ADWI and the intake of other nutrients or between ADWI and the concentration of nutrients in the diet; however, N concentration in the diet approached significance (r = 0.25; P < 0.10).
|
and 5. The ability to predict ADWI was best accomplished using N, Na, and Cl. Using the daily intake of these 3 nutrients resulted in an R2 of 0.26 (P < 0.004), whereas a combination of their concentration in the diet and ADFI resulted in an R2 of 0.1 (P < 0.003). The most effective prediction of ADWI was achieved when ADFI was used in combination with daily intakes of N, Ca, P, K, and Cl (R2 = 0.33; P < 0.017). In any event, the most comprehensive equations presented in Tables 4 and 5 achieved only modest success in predicting ad libitum water intake.
|
|
).
|
). Feed intake was greatest on the LoP diet (P < 0.05).
|
). However, N retention (g/d) was equal across treatments, as was the estimated protein deposition rate. Percent N digestibility was reduced in the ExcM diet, and the percent N retention was reduced in the ExcP diet (P < 0.05). Otherwise, reducing dietary CP level did not affect N excretion, retention, or digestibility values, when expressed as a percentage of intake.
|
) compared with the other treatments. Pigs on the HiP diet consumed the least amount of Ca and Na (P < 0.05). Chloride intake was lowest in the pigs fed the HiP and ExcP diets (P < 0.05). Potassium intake was proportional to N intake (P < 0.05). Total excretion of Ca and P was greatest in the ExcM-fed pigs, and total excretion of Na and Cl was greatest in the ExcM and LoP-fed pigs. Despite the high rates of excretion, the pigs fed the ExcM diet retained the greatest quantity of Ca, P, and Na, compared with all other treatments (P < 0.05), when expressed on a grams per day basis. Pigs on the LoP diet had the least retention of Ca and K, whereas pigs on the LoP and HiP diets had the least retention of Na (P < 0.05). The retention of P was similar on the LoP and HiP diets but greater on the ExcM diet and lesser on the ExcP diet.
|
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
). For example, under limit-feeding conditions, the most desirable when evaluating nutrient utilization, pigs consume excessive and highly variable quantities of water (Yang et al., 1981). In a preliminary experiment in which feed intake was restricted (Shaw, 2003), the SE for the water:feed ratio was 0.37; in this experiment, with ad libitum access to feed, the SE was reduced to 0.15. Excess water intake, often referred to as hunger-induced polydypsia, was also reduced by ad libitum feeding; in the preliminary study, the water:feed ratio was 3.57, compared with the 2.69 reported herein. Thus, pigs were allowed ad libitum access to feed in this experiment. Although this is not the best circumstance for evaluating dietary effects on nutrient utilization, it certainly is the best platform for the evaluation of water utilization. Because this was the primary objective of the experiment, the ad libitum circumstance was employed in the current experiment.
The boredom associated with individual housing can play a role in water usage (Patience et al., 1987b; Fraser et al., 1990). In this experiment, housing animals in individual metabolism crates may have increased water intake somewhat over group-housed conditions. However, we believe the impact was very small because the design of the metabolism crates employed in this study allowed normal animal movement, and thus normal feed intake, compared with the more commonly used metabolism crates that severely restrict animal movement. Additionally, the water in-take:feed intake ratio reported herein was 2.69:1, similar to the ratio of 2.43 reported by Li et al. (2005) for growing pigs housed in group pens under thermoneutral conditions.
To consider more than simple drinking water intake and urine output, direct measurements or estimates of other components of the pig’s water balance were included. Because of the difficulties inherent in such an analysis under conditions other than respiratory chambers, respiratory losses were not measured and metabolic water production was estimated indirectly. However, use of respiratory chambers introduce other errors in estimating normal water utilization, so the current experiment model was considered best suited to achieve the objectives of this experiment. Estimating those components of water balance not measured directly served to illustrate the importance and relative contribution of the other components that might otherwise be ignored or forgotten. Brooks and Carpenter (1990) undertook a very detailed analysis of the production of metabolic water, as well as quantifying other components of the water balance. In our study, drinking water represented 83% of total measured intake, with metabolic water adding 14% and feed water 3%. Although drinking water is clearly the predominant source, metabolic water also contributed substantially to the water balance in these pigs and therefore should not be ignored in commercial practice.
The portion of total measured water output recorded as urine averaged 83%. Because urine output will be most directly related to water intake, at least under conditions of normal health and environment, one might reasonably assume that as water intake increases, urine output will similarly rise.
One of the fundamental assumptions associated with diet composition affecting water intake involves the biological drive to maintain constant, or at least relatively constant, urine osmolality. If the composition of the diet increases the need to eliminate greater quantities of metabolic by- and end products plus unused nutrients, drinking water requirements will only increase if the pig must maintain a constant or near-constant urine osmolality. If excreting more concentrated urine is acceptable, the quantity of drinking water need not increase, or need not increase in proportion to the need for nutrient excretion. The need to maintain constant urine osmolality is not absolute but relative; although there is a limit to the range of possible urine osmolality, the solute concentration in urine certainly can and does vary. Pond and Houpt (1978) reported that urine osmolality in the pig may range from 253 to 994 mOsm/kg. Schiavon and Emmans (2000) established a water requirement for urinary excretion of about 2.05 and 3.40 L of water/osmol of urea and electrolytes, respectively, but did not define the range about the mean that might be observed under diverse physiological and nutritional circumstances. The fluctuation in urine osmolality contrasts the very tight control of osmolality of the plasma. As an example, McLeese et al. (1992) reported that the osmolality of swine blood was kept fairly constant at 1.0 mOsm/kg.
Effect of Dietary Protein Level on Water Utilization Patterns
The minimum requirement for drinking water in growing pigs, according to Yang et al. (1981) is a water:feed ratio of 1.6:1. The water:feed ratios reported herein ranged from 2.5:1 to 3.1:1. Though slightly greater, these observed ratios are comparable to those reported by NRC (1998), which states that when given free access to feed, pigs will drink water in a 2.0:1 to 2.5:1 ratio to feed intake.
Whereas some studies have shown an increase in drinking water associated with increasing dietary CP levels (Suzuki et al., 1998), others observed a lack of effect (Jongbloed et al., 1997). In our experiment, excess protein, but not high protein, tended to increase consumption (P < 0.10) and significantly increased the water:feed ratio (P < 0.05). Also, including daily N intake or dietary N concentration in the stepwise regression improved the prediction of ADWI. These data are supported by the work of Le Bellego and Noblet (2002) in which weaned pigs given a low protein diet tended to consume less water and excrete less urine compared with pigs fed a diet with a greater CP concentration.
Water intake was correlated with dietary CP level in the studies by Pfeiffer et al. (1995) and Suzuki et al. (1998); however, these experiments contained significant confounding factors. Pfeiffer et al. (1995) reported that increased dietary CP concentration resulted in increased water intake, but in their experiments, extreme changes in dietary protein concentration were imposed on the pigs: 18.5 vs. 25.0% and 10.4 vs. 20.4% in Exp. 1 and 2, respectively. The experimental diets were not balanced nutritionally, so the comparison of CP level was confounded by amino acid adequacy. For example, the daily lysine intake in their Exp. 2 ranged from 4.6 to 12.9 g/d. It is therefore impossible to conclude from their data whether water intake is driven by dietary protein concentration or dietary protein adequacy. Our data suggest that excess protein, rather than simply an increase in protein level, increases ADWI.
Effect of Dietary Mineral Level on Water Utilization Patterns
We were surprised by the lack of response in water intake to elevated dietary mineral content. Elevated mineral intake tended to increase fecal water excretion, but not urine output. Perhaps more surprising was the absence of an effect on urine osmolality. The absence of an effect of mineral intake on water output may be explained by the balance data. The excess mineral diet increased the retention of Ca, P, Na, and Cl. Furthermore, the excess mineral diet increased ADG, but not estimated protein gain. It is therefore possible that, in such a short-term study, the excess mineral diet resulted in mineral retention rather than mineral excretion. A longer-term experiment may have yielded different results in this regard.
A lack of response in water intake to diets differing in Na was also observed by Alcantara et al. (1980). It is also possible that the amount by which mineral levels were increased to attain the levels of the ExcM diets was not great enough to elicit an overall treatment effect on drinking water intake. A 450% increase in dietary salt load was required to produce an increase in water demand of only 16%/kg of diet (Hagsten and Perry, 1976). Similarly, in a study by Seynaeve et al. (1996), water intake increased in sows fed a diet containing 0.85% NaCl, compared with that of sows receiving a diet with 0.1% NaCl.
It is clear from the regression analysis that water consumption is influenced by factors other than diet composition. No equation predicted ADWI with an R2 greater than 0.33, indicating that another, or many other, factors are involved.
Effect of Diet on Pig Performance and Nitrogen Balance
The observed improvement in ADG in pigs on the ExcP diet relative to the HiP diet was mirrored by increased G:F. Though fecal and urinary N excretion was greater, overall N retention was unaffected by the increased dietary CP level, suggesting that protein deposition was equal on the HiP and ExcP diets. The tendency for increased water intake on the ExcP diet compared with the other protein levels, combined with no change in N retention, could suggest that ExcP merely increases water retention. Although this is not completely supported by the water balance data, there was a tendency for increased water retention in the pigs fed the ExcP diet.
Effect of Diet on Mineral Balance
The amount of Ca and P excreted in the feces was greater with the ExcM diet than the control (HiP) diet, yet P retention as a percentage of intake was not affected by diet, whereas percentage retention of Ca varied with diet. This is again perhaps related to incomplete adaptation to the experimental diets. Pointillart and Fontaine (1986) stated that a changed dietary supply of Ca results in a longer adaptation of the absorption of Ca than is the case for P. This may help to explain the differences in Ca retention vs. the similarity observed in P retention, among dietary treatments.
Retention values for Na, K, and Cl were relatively lesser than those associated with Ca and P. Swine excrete Na, K, and Cl at a rate of 70 to 95% of intake (Kornegay and Harper, 1998). Studies have shown increases in urinary Na and Cl concentration in response to increasing dietary levels of these minerals (Hagsten and Perry, 1976; Seynaeve et al., 1996). The experimental diets were formulated to meet or exceed NRC (1998) requirements; diet analyses confirmed that mineral levels were in excess of the recommended minimums. Similar to the results obtained for Ca and P, retention of Na, K, and Cl, when expressed as a percentage of intake, was less overall when pigs were given free access to feed vs. when pigs were limit-fed. The negative retention values obtained for some minerals may be a result of an inadequate period of adaptation of pigs to the experimental diets as discussed previously.
|
IMPLICATIONS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
|
Footnotes |
|---|
2 Current address: Phibro Animal Health, Calgary, AB T2W 2L9.
3 Corresponding author: john.patience{at}usask.ca
Received for publication November 25, 2005. Accepted for publication June 28, 2006.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. J. Kerr, C. J. Ziemer, T. E. Weber, S. L. Trabue, B. L. Bearson, G. C. Shurson, and M. H. Whitney Comparative sulfur analysis using thermal combustion or inductively coupled plasma methodology and mineral composition of common livestock feedstuffs J Anim Sci, September 1, 2008; 86(9): 2377 - 2384. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
