By Kim, B G Lindemann, M D
ABSTRACT A knowledge of nutrient requirements is essential to allow fuller expression of genotypic ability while minimizing unwanted economic expense in swine production. The NRC publications were developed to provide this information and have been widely used in academia and in the swine feed industry. However, the accuracy of nutrient requirements, especially minerals and vitamins, is often a subject of debate. Thus, the objective of the current paper was to review the mineral and vitamin requirements in the NRC publications. To accomplish this, the mineral and vitamin requirements in the NRC publications from 1944 to 1998 were restructured and summarized. Initial nutrient requirement estimates and changes over time were tabulated. The number of citations supporting the relevant nutrient requirement was counted as a measure of the reliability of the requirement estimation. The contemporaneity of supporting studies was measured by counting the citations by decades. In summary, the number of minerals and vitamins for which a requirement is provided has increased from the initial publication. However, many of these requirement estimates have not changed much over the years. Recent citations for many minerals and vitamins were lacking, especially for starting pigs and reproducing sows. These research gaps in the mineral and vitamin nutrition of swine may illustrate unknown limitations on health and performance and may be opportunities for research.
Key words: nutrient requirements, NRC, pigs, minerals, vitamins
INTRODUCTION
In livestock production, an accurate understanding of dietary nutrient requirements is important for maintaining the normal health of animals, for maximizing the productive performance of animals, for reducing environmental pollution, and for maximizing producer profits. Over the past 6 decades, the NRC has published 10 editions of the nutrient requirements of swine (1944, 1950, 1953, 1959, 1964, 1968, 1973, 1979, 1988, 1998), summarizing available information related to nutrient requirements and establishing those requirements for pigs at several production stages. Although similar publications are also available in many countries (ARC, 1981; SCA, 1987; CVB, 2004; INRA, 1984), the NRC publications are undoubtedly the most frequently cited in scientific papers. And, although they are minimum requirements, they are the reference values used before the addition of safety margins in much of the swine feed industry worldwide.
The NRC
The original NRC (1944) was a relatively brief document. It stated that at the turn of the century there was an understanding of the caloric and protein needs of pigs (and to some degree an understanding of the Ca and P needs), but with research conducted since then there was a need to assemble a current description of the nutrient needs. The publication included 73 citations (with a large number being from experiment station bulletins) and presented the suggested nutrient allowances (as opposed to nutrient requirements). The publication had 3.5 pages of text and 1 table each of nutrient allowances (14 nutrients), clinical symptoms of deficiency, partial composition of feedstuffs, and sample diets (to illustrate nutritional principles). Pictures were included of vitamin A, thiamin, Ca, and pantothenic acid deficiencies. The BW range for growing pigs was from 50 to 250 pounds (23 to 114 kg).
The second NRC (1950) was double in length and included 3 amino acids in the nutrient allowances table (Trp, Lys, and Met). The third NRC (1953) is most notable for the title change to nutrient ‘requirements’ rather than nutrient ‘allowances.’ It also discussed antibiotics in relation to vitamin B12 and the Animal Protein Factor. The BW range of growing pigs was widened to 25 to 250 pounds (11 to 114 kg). This BW range was further changed in the fourth NRC (1959) to 10 to 200 pounds (5 to 91 kg). Feed composition tables were omitted from the fourth edition and citations began to be grouped by nutrient or production stage for ease of assessment by readers.
Feed composition tables were returned in the fifth edition (NRC, 1964) and the weight range was changed again, this time to 10 to 225 pounds (5 to 102 kg). Discussion was included relative to the avoirdupois system of weights (pounds and tons) compared with the metric system, but there were no changes made in the nutrient tables. However, in the sixth edition (NRC, 1968) the pig weights were changed to metric and the weight range used was 5 to 100 kg. This edition also contained much more extensive text and feed composition tables.
The seventh edition (NRC, 1973) maintained the same weight ranges but began to include more extensive footnoting of tables to describe the assumptions of the requirement estimate tables, most notably that the estimates were for a corn-based diet. This continued with the eighth edition (NRC, 1979), but the assumption noted was that the estimates were for a grain-soybean meal diet. The weight range for which estimates were provided in NRC (1979) was 1 to 100 kg. This was the first edition in which pictures of selected deficiency symptoms were not provided.
In the NRC (1988), the introduction states that the edition is substantially different than previous editions, that “the major change is the documentation of scientific papers used to determine the nutrient requirements.” It further highlights the amino acid focus (and diminished attention to CP) and the introduction of ileal digestibilities for 18 of the most common feedstuffs. Bioavailability of P in 20 of the most common feedstuffs is provided. Whereas this was the most extensive NRC publication to date, it clearly stated that “There is conflicting, incomplete, or no information for several nutrients at different stages of the life cycle . . . This is particularly true for many of the vitamins and trace minerals.”
The most recent edition of NRC (1998) uses a mathematical modeling approach to predict nutrient needs for swine. A model for the growth of pigs estimates energy and amino acid requirements based on the gender, rate of lean gain, and various environmental factors. Models for gestating and lactating sows were also developed. These excellent models are very flexible and dynamic for predicting energy and amino acid requirements. However, a modeling approach was not used in estimating mineral requirements mainly due to the insufficient data for accurate model development; thus, Cromwell and Baer (2005) note that all the mineral requirements were estimated based on empirical data from research papers. The estimation of vitamin requirements also did not employ a modeling approach.
The lack of studies for some minerals and vitamins [as stated explicitly in NRC (1988) and implicitiy in NRC (1998) by its inability to develop models] does not mean that these nutrients are insignificant. They are extremely important to overall health and productivity of animals. Although dietary energy and protein constitute the greatest economic cost within diets, it is minerals and vitamins that are most often evaluated with regard to the critical health aspect of immunocompetence and, because almost all metabolic reactions require one or more of them as cofactors, they dictate the efficiency of utilization of the energy and protein that is provided. Thus the awareness of accurate requirement estimates of these 2 nutrient classes will improve the nutritional well-being of pigs. Additionally, in some measure, the number of studies supporting a specific nutrient requirement estimate should represent the degree of scientific reliability of the requirement estimate. A lack of supporting studies may indicate future research directions or opportunities. Therefore, the objective of the cunent review was to tabulate the requirement estimates for minerals and vitamins for various production stages of swine as well as the relevant number of studies supporting those requirement estimates in NRC publications published from 1944 to 1998.
REVIEW PROCEDURE
Nutrient Requirement Estimates
Mineral and vitamin requirement estimates provided in the 10 NRC publications (1944 to 1988) were reorganized and summarized. The production stages of swine were categorized as starting pigs (BW of 3 to 20 kg), grow-finishing pigs (BW of 20 to 120 kg), and reproducing sows (gestating and lactating). All avoirdupois units were converted to metric units. When the nutrient requirements were expressed only as a daily intake amount in the particular NRC tables, they were converted to a concentration unit basis [%, g/kg, mg/kg, or IU/kg; following the units in the tenth revised edition of NRC (1998)] based on the estimated daily feed intake in that same publication. When more than one requirement estimate value was given for the BW ranges within the starting and grow-finishing stages in the NRC publications, the nutrient requirement in this review was provided as a range. Also for reproducing sows, when the nutrient requirements for gestating and lactating sows (or gilts) were not the same, the values were expressed as a range.
Citation Count of Cited References
All references cited in the mineral and vitamin sections in each edition of the NRC were pooled, and duplicates were removed for the citation counting. Each reference was then categorized electronically by its publication year, the NRC publications in which it was cited, the swine production stage evaluated, and the nutrients evaluated. The production stages were starting pigs, grow- finishing pigs, reproducing sows, as well as nonswine publications. When the reference title used the terms of starting, weaning, weanling, weaned, baby, young, or suckling, the reference was regarded as a study for starting pigs. The references containing reproduce, pregnant, gravid, parental, maternal, fertility, gestation, or lactation in the title were considered as reproducing sow citations. All other citations using pigs in their studies were regarded as grow-finishing stage papers. As stated, some of the citations used nonswine animal models, and these are not presented herein.
Although most of the research publications studied a single nutrient, some studies investigated more than one nutrient. In this case, the citation was counted for all the nutrients listed in the title. For example, Cromwell et al. (1970) studied both Ca and P, and this citation was counted for both.
The mineral and vitamin terminology in NRC (1998) is used as a standard in this review. The mineral names were relatively consistent in the citation titles, but vitamin names were different depending on the authors and the publication years. Thus, various vitamin names were classified as follows: retinol-, caroten-, cryptozan-, and A-avitimin as vitamin A; calciferol as vitamin D; tocopher- and tocotrien- as vitamin E; phylloquinone, menaquinone, menadione, naphthoquinone, and prothrombin factor as vitamin K; vitamin B! and aneurin as thiamin; vitamin B2 and lactoflavin as riboflavin; vitamin B3, nicotinamide, and nicotinic acid as niacin; vitamin B5 and pantothe- as pantothenic acid; pyridoxine and pyridoxal- as vitamin B6; -cobalamin and dibencozide as vitamin B12; and pteroylglutamic acid and vitamin B9 as folacin. When the topic nutrient of a citation was not stated in the title but the citation was clearly classified by the section of the bibliography from which it was obtained, the section title was regarded as a topic nutrient of the citation. For example, a study by Muhrer et al. (1970) was cited in the seventh edition of NRC (1973) without any specific nutrient in the title, but this citation was counted as a vitamin K related paper following the section in which it was found in the bibliography.
Although the above method of electronic counting can give a very quick breakout of the citations in various categories (as presented in Tables 4 to 6), it can potentially overestimate the amount of research associated with a particular nutrient requirement estimate. For example, a citation may be of a textbook that describes the metabolic function of a particular nutrient; although this citation has value in the text, it obviously would not be used in the establishment of an absolute requirement estimate. Further, in order to be used to establish a requirement estimate, at least 3 levels of supplementation would be needed; the use of 2 levels allows no estimate of where a breakpoint in the supplementation would occur. Therefore, to provide a more conservative estimate of the number of citations associated with the requirement estimates in the most recent NRC (1998), a manual count and grouping of the citations was conducted to develop Table 7. Examples of the decision making process of our effort are provided in Tables 8 and 9. Although this manual count is meant to be more conservative than the electronic count to more accurately reflect the number of studies that may have contributed to the actual requirement estimate listed, decisions about how to classify a citation simply by its title can be arbitrary. Appraisal of Tables 8 and 9 should demonstrate a relatively liberal classification to avoid the possibility of being unduly critical of the amount of data that may have been used in developing the requirement estimate. Finally, during this manual counting process, an actual identification of citations related to a previously neglected category of summarization (i.e., boars) was made and is provided as Table 10.
RESULTS AND DISCUSSION
Mineral and Vitamin Requirement Estimates of Swine
The mineral and vitamin requirements of swine in the NRC publications (1944 to 1998) were summarized for starting pigs (Table 1), growfinishing pigs (Table 2), and reproducing sows (Table 3). The NRC has evolved both by adding new requirement estimates for more minerals and vitamins, by updating the requirement values, or both.
In the first edition (NRC, 1944), the requirements for Ca, P, Na, and K were provided. In this edition, the essentiality of other minerals including Mg, Cu, I, Fe, Mn, and Zn were mentioned in the text, but the quantitative requirement values were not provided in the table due to the lack of data for requirement estimation. This remained essentially the same for the second edition (NRC, 1950) with the exception that a value for salt was provided instead of values for Na and K. In the third edition (NRC, 1953), quantitative values for Cu, I, Fe, and Mn were listed (Tables 1, 2, and 3) in a separate table for all swine without distinguishing among production stages. However, as stated earlier, they were provided not as minimum requirements but as tentative recommendations. In the next edition (NRC, 1959), the requirements of Mg and Zn were listed, and Se first appeared textually as an essential nutrient. The quantitative requirement of Se was first shown in the fifth edition (NRC, 1964). The requirement estimate for available P was added in the ninth edition (NRC, 1988). Chromium was presented textually as an essential nutrient, but the requirement was not estimated, in the tenth edition (NRC, 1998). For reproducing sows, the requirement estimates for most of the microminerals were not specifically available until the eighth edition (NRC, 1979; Table 3).
Table 1. Dietary mineral and vitamin requirement estimates for starting (3 to 20 kg) pigs from the 10 NRC publications1
Table 2. Dietary mineral and vitamin requirement estimates for grow-fishing (20 to 120 kg) pigs from the 10 NRC publications1
Table 3. Dietary mineral and vitamin requirement estimates for reproducing sows from the 10 NRC publications1
The requirement estimates of vitamins including vitamin A, vitamin D, thiamin, riboflavin, niacin, pantothenic acid, and vitamin B6 were given in the first edition (NRC, 1944). The requirement values of vitamin B12 and choline (only for starter pigs; Table 1) were added, and the necessity of biotin was discussed in the third edition (NRC, 1953). In the same edition, vitamin E was described as an essential element, and its quantitative requirement value was established in the seventh edition (NRC, 1973). The requirement estimates for vitamin K, biotin, choline, and folacin for grow-finishing swine were added in the eighth edition (NRC, 1979; Table 2).
Table 4. Number of references cited specifically related to the listed nutrients in the 10 NRC publications1
The requirement estimates for many minerals and vitamins have not changed much over the years and, in some cases, the requirement estimates for minerals and vitamins have decreased. Manganese is a good example of requirement decrement. The provisional recommendation value of 40 mg/kg (Table 2) was provided in the third edition (NRC, 1953) and this number was changed to 20 mg/kg in the sixth edition (NRC, 1968). In the eighth edition (NRC, 1979), the requirement estimate was revised again as 2 mg/kg. The revision of requirement estimates generally reflects new research data. Alternately, relatively consistent requirements of some nutrients over the 5 decades of publications (NRC 1944 to 1998) would imply that 1) the initially established value was close to the real requirement with only minor adjustments required following further studies, or 2) no additional information about the requirement of a specific nutrient was subsequently available. In the latter case, more studies may be required to improve the accuracy of the requirement estimates. The citation counting for each nutrient discussed in the next section may identify some of these gaps in mineral and vitamin requirement related studies.
Table 5. Number of references cited specifically related to the listed nurtrients for starting, grow-finishing, and reproducing swine in the 10 NRC publications1
Number of References Cited
Table 4 provides the number of references cited related to a specific mineral or vitamin in each edition of NRC (1944 to 1998). Obviously, the number of citations in the NRC publications has increased over the 5 decades. However, this increase varied among the nutrients. As a cautionary note, the number of citations in this table was counted based only on the topic nutrients in the reference title regardless of the species or age of animals used in the experiments. Subsequently, the number of citations was divided into 3 different production stages of starting pigs, grow-finishing pigs, and reproducing sows (Table 5). The number of citations was highest in the grow-finishing pigs and lowest in reproducing sows for all the minerals and most of vitamins. For starting pigs and reproducing sows, no reference was cited for some nutrients even in the most recent edition of NRC (1998). As mentioned in a recent publication (Cromwell and Baer, 2005), many mineral requirement estimates for sows in NRC (1998) were based on the information from growing pigs. This fact would surely be the same for vitamin requirement estimation.
Table 6. Number of references by decades specifically related to the listed nutrients in NRC (1998)1
In order to identify the quantity of citations and their contemporaneity, the number of references in the most recent NRC (1998) related to specific nutrients was counted by decades in which the references were published (Table 6). The nutrients for which citations were more than 50 include Ca, P, Cu, Fe, Se, Zn, and vitamin E, implying extensive research conducted for these nutrients. In contrast, for Mg, I, Mn, vitamin K, and thiamin, less than 15 citations were listed in the bibliography of this most recent NRC edition. Considering that not all the citations counted here may be highly qualified or applicable to the requirement estimates, the information about these nutrients would be very limited. Moreover, less than 6 citations were published in the 1980s and ’90s for Mg, I, Mn, vitamin K, thiamin, riboflavin, niacin, pantothenic acid, vitamin B6, and vitamin B12. The lack of relatively recent data implies a potential inaccuracy of requirement estimates if there have been metabolic or genetic changes in current pigs or changes in feed intake resulting from changes in swine production management. The requirement estimates for most of these nutrients lacking recent studies have no major updates (Tables 1, 2, and 3). In other words, the requirement estimates were established based on studies conducted at least 25 yr ago. The nutrient requirements for modern types of animals may not be the same as before. This potential inaccuracy of requirement estimate is demonstrated in a recent study for dietary B vitamins (riboflavin, niacin, pantothenic acid, vitamin B12, and folacin). These vitamins were tested in moderate- and high-lean type pigs by Stahly et al. (2007). The needed dietary concentration of these B vitamins for optimum feed efficiency was reported to be 2.7 times the NRC (1998) estimates in moderate lean type pigs and greater than 4.7 times the NRC (1998) estimates in high lean type pigs. A greater need for at least one of these vitamins was clearly shown, therefore, but the requirement for any single vitamin of the 5 evaluated was unavailable from this study because the dietary concentrations of all 5 B vitamins were changed together across the treatments. Table 7. Number of references by decades specifically related to the listed nutrients as determined by manual count in NRC (1998)1
Table 8. Illustration of the decision process for mineral citation counting in the NRC (1998) for Table 7
Table 9. Illustration of the decision process for vitamin citation counting in the NRC (1998) for Table 7
Further information of the contemporaneity and quantity of citations is provided by dividing the decade counting into the production stages of swine and by conducting an actual manual count of the citations. This gives an even better understanding of current research gaps for minerals and vitamins (Table 7). The number of publications for grow-finishing pigs was relatively high. However, again as a cautionary note, some of the counted citations were nonrefereed articles or abstracts, which may diminish the value of its contribution to a requirement estimate. For reproducing sows, an extremely limited number of citations, or none, are supporting some of the mineral and vitamin requirement estimates. As aforementioned, many of the requirement estimates of minerals and vitamins for sows were established by the summation and interpolation of the data from growing pigs. This estimation method was perhaps the best available to set the sow requirement estimate without enough applicable data. But the accuracy of requirement estimates generated in that manner is still questionable. Certainly this does not result in the provision of equal amounts of nutrients per unit of body mass to that of the younger pig and may result in long term tissue depletion of some nutrients.
Finally, with regard to the boar (Table 10), the dearth of information is clearly evident. A degree of information related to Ca and P needs has been generated, but essentially no information relative to actual reproductive function is available.
IMPLICATIONS
This review has demonstrated that the requirement estimates for swine have, in many cases, changed little over the decades. However, the individual nutrient estimates do vary markedly with regard to the amount of direct research that was used to develop those estimates and with regard to when it was conducted. In discussing or deciding the nutrient allowances for some nutrients for some stages of production, then, a degree of caution is warranted. An openness to the practical experience of others that have utilized common genetics and feedstuffs to the situation being discussed may be prudent. For continued improvement in swine nutrition and, hence, swine production, further evaluations and updates are required for some nutrients in starting and grow-finishing pigs, for most nutrients in reproducing sows, and certainly for boars.
Table 10. Citations related to minerals and vitamins for boars
1This manuscript is published by the Kentucky Agricultural Exp. St. as paper no. 07-07-071.
LITERATURE CITED
ARC. 1981. Agricultural Research Council: The Nutrient Requirements of Pigs. Commonwealth Agricultural Bureaux, Farnham Royal, London, UK.
Cromwell, G. L., and C. K. Baer. 2005. Setting the NRC standards for minerals – were we right? Page 1 in Redefining Mineral Nutrition. J. A. Taylor-Pickard and L. A. Tucker, ed. Nottingham University Press, Nottingham, UK.
Cromwell, G. L., V. W. Hays, C. H. Chaney, and J. R. Overfield. 1970. Effects of dietary phosphorus and calcium level on performance, bone mineralization and carcass characteristics of swine. J. Anim. Sci. 30:519.
CVB. 2004. Veevoedertabel. Centraal Veevoederbureau, Lelystad, The Netherlands.
INRA. 1984. L’alimentation des animaux monogastriques: pore, lapin, volailles. Institut national de la recherche agronomique, Paris, France.
Lindemann, M. D., and B. G. Kim. 2006. Recent advances in sow reproductive function. Page 25 in Nutritional Approaches to Arresting the Decline in Fertility of Pigs and Poultry. J. A. Taylor- Pickard and L. Nollet, ed. Wageningen Academic Publishers, The Netherlands.
Muhrer, M. E., R. G. Cooper, C. N. Cornell, and R. D. Thomas. 1970. Diet related hemorrhagic syndrome in swine. J. Anim. Sci. 31:1025. (Abstr.)
NRC. 1944. Recommended Nutrient Allowances for Swine. Natl. Acad. Press, Washington, DC.
NRC. 1950. Recommended Nutrient Allowances for Swine. 2nd rev. ed. Natl. Acad. Press, Washington, DC.
NRC. 1953. Nutrient Requirements for Swine. 3rd rev. ed. Natl. Acad. Press, Washington, DC.
NRC. 1959. Nutrient Requirements of Swine. 4th rev. ed. Natl. Acad. Press, Washington, DC.
NRC. 1964. Nutrient Requirements of Swine. 5th rev. ed. Natl. Acad. Press, Washington, DC.
NRC. 1968. Nutrient Requirements of Swine. 6th rev. ed. Natl. Acad. Press, Washington, DC.
NRC. 1973. Nutrient Requirements of Swine. 7th rev. ed. Natl. Acad. Press, Washington, DC.
NRC. 1979. Nutrient Requirements of Swine. 8th rev. ed. Natl. Acad. Press, Washington, DC.
NRC. 1988. Nutrient Requirements of Swine. 9th rev. ed. Natl. Acad. Press, Washington, DC.
NRC. 1998. Nutrient Requirements of Swine. 10th rev. ed. Natl. Acad. Press, Washington, DC.
SCA. 1987. Feeding Standards for Australian Livestock. Pigs. Commonwealth Scientific and Industrial Research Organisation, East Melbourne, Australia.
Stahly, T. S., N. H. Williams, T. R. Lutz, R. C. Ewan, and S. G. Swenson. 2007. Dietary B vitamin needs of strains of pigs with high and moderate lean growth. J. Anim. Sci. 85:188.
B. C. Kim and M. D. Lindemann2
Department of Animal and Food Sciences, University of Kentucky, Lexington 40546-0215
2 Corresponding author: [email protected]
Copyright American Registry of Professional Animal Scientists Dec 2007
(c) 2007 Professional Animal Scientist. Provided by ProQuest Information and Learning. All rights Reserved.
Comments