J. Anim Sci. 2008. 86:364-369. doi:10.2527/jas.2006-601
© 2008 American Society of Animal Science
An estimate of the methionine requirement and its variability in growing pigs using the indicator amino acid oxidation technique1
S. Moehn*,
A. K. Shoveller*,2,
M. Rademacher
and
R. O. Ball*,3
* Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Alberta, Canada T6G 2P5; and
Degussa GmbH Feed Additives, 63457 Hanau, Germany.
 |
Abstract
|
|---|
Although AA requirements for the mean of a population of growing pigs have been established using traditional methods, there are no estimates of the variability within the population and whether this variation differs among AA. With the increased use of supplemental Lys in pig diets, there will be an increased need to supplement Met, commonly the second or third limiting AA in corn-soybean diets. The indicator AA oxidation method allows repeated measurements in a short period of time so that the AA requirement can be determined for individual pigs at a similar physiological stage. The objective of this study was to determine the mean Met requirement in individual gilts and to estimate the related variability. Six individually housed female pigs (initial BW = 8.8 kg, SD 1.5) each received diets providing 6 levels of DL-Met. The isonitrogenous and isoenergetic diets contained 0.187, 0.250, 0.290, 0.320, 0.350, and 0.377% Met (analyzed, as-fed basis). Cysteine (0.48%) and Lys (1.44%) concentrations were similar for all diets. Pigs were adapted for 6 d to the basal corn-soybean meal diet (0.187% Met), which was offered at 95 g/kg0.75 of BW to ensure complete consumption of the test diets. During 4-h oxidation studies, 313.4 kBq, (SD 35.6) of L-[1-14C]Phe was mixed with each of 8 half-hourly meals, and expired CO2 was collected. The breakpoint in Phe oxidation, representing the Met requirement, and its variability, was determined using 2-phase linear regression. Phenylalanine oxidation decreased as the Met content increased from 0.187 to 0.29%. Phenylalanine oxidation was not different (P > 0.2) for diets ranging from 0.320 to 0.377% Met. The dietary Met requirement varied from 0.320 to 0.373% for individual pigs. The mean Met requirement for individual pigs was determined to be 0.340% of diet (SD = 0.024%, CV= 7.1%), with 0.340, 0.364, and 0.388% covering the requirement of 50, 66, and 95% of the population, respectively. The present mean population estimate was similar to the recommended dietary Met concentration of 0.325% for pigs of this BW and feed intake. To maximize profitability, Met levels in starter pig diets should be determined, depending on the cost of crystalline Met and the fraction of the population whose requirement is to be met.
Key Words: gilt • indicator amino acid oxidation • methionine • requirement • variability
|
INTRODUCTION
|
|---|
Commonly, Met is the second or third limiting dietary AA, and because of the increased use of lower-protein swine diets that require addition of crystalline Lys, there is greater need for Met supplementation. Conventional methods of requirement determination provide an estimate of the mean requirement of a population; this means that the requirement of 50% of all pigs may not be met. The variability of AA requirements must be known to supply AA at the economic optimum (i.e., discounting the cost of additional AA supplementation against animal performance gained). Traditional methods, like growth rate or N balance assays, are unsuitable to determine requirements of individual growing animals. Testing 6 or 7 dietary AA levels within an individual growing animal would require at least 2 mo so that the requirement would change substantially during the experiment. The indicator AA oxidation technique (IAAO), which was developed in pigs (Kim and Bayley, 1983
), has been widely used to determine AA requirements in pigs and humans (Brunton et al., 1998; Pencharz and Ball, 2003) and poultry (Coleman et al., 2003), although the IAAO requires specialized equipment and expensive material, such as isotope and equipment for oxidation studies and analyses. The IAAO allows measurements to be made after an adaptation period of 2 d (Moehn et al., 2004) so that sufficient AA intake can be tested in a 2-wk experiment without the animal changing its physiological state. By controlling animal BW, feed intake, and environmental conditions, the variability in requirements can be reduced to factors associated with the genotype of the individual animal. Therefore, the goal of this experiment was to use the IAAO and oral isotope dosing (Moehn et al., 2005) to determine the minimum Met requirement in growing pigs fed a corn-soybean meal-whey powder-based diet and derive an estimate of the interanimal variation and hence the associated overall variability.
|
MATERIALS AND METHODS
|
|---|
Animals and Study Protocol
All procedures used in this study were approved by the Faculty of Agriculture, Forestry and Home Economics’ Animal Policy and Welfare Committee of the University of Alberta. Six Genex Manor hybrid F1 gilts (Genex Swine Group, Regina, Saskatchewan, Canada; initial BW = 7 kg) were initially kept in pairs to minimize stress and were allowed to consume the experimental basal diet during this 7-d period on an ad libitum basis. Gilts were separated and housed individually after the initial acclimatization period and were then adapted over a 6-d period to the basal corn-soybean diet, which was offered at 95 g/kg0.75 of BW to ensure complete consumption. Gilts were allowed to consume water ad libitum from automatic drinkers.
After the 6-d adaptation period, gilts were randomly allocated to 1 of 6 test diet concentrations of Met. Two days later, the first IAAO study was conducted. After each oxidation study, gilts received the next test diet and were adapted again for 2 d (Moehn et al., 2004) before the subsequent IAAO study was conducted. This procedure was repeated until all gilts had received each diet, thereby ensuring that each diet was represented at each time point. At the end of the experiment, the pigs were euthanized by an injection of pentobarbital.
Diets and Feeding
The pigs were fed twice daily (0700 and 1700), except for the oxidation study days, when they received half the daily ration divided in 16 half-hourly meals beginning 30 min before commencing the oxidation study. The remaining half of the daily ration was fed in the evening as on other days. This frequent feeding protocol ensured that pigs remained in the fed state, which resulted in a more stable isotopic plateau. Feed intake was restricted to 95 g/kg0.75 of BW to ensure complete consumption of the diet and the isotope at each feeding. The feed was allocated based on the BW of each gilt, which was determined on the morning of oxidation study days and after an overnight fast from 1700 to 0700. Any feed not eaten was collected, dried, and weighed to obtain net feed intake.
Each animal received, in random order, diets providing 1 of the 6 Met concentrations, with a constant intake of Cys (0.48% of diet analyzed, as-fed basis) and Lys (1.46% of diet analyzed, as-fed basis). Cysteine was offered at over 100% of its requirement to minimize the amount of Met that was metabolized to Cys via the transsulfuration pathway. Diets were based on corn-soybean meal and formulated to be isonitrogenous and isoenergetic (Table 1). Each pig received, in random order, diets providing Met at 0.187, 0.250, 0.290, 0.320, 0.350, and 0.377% Met, as analyzed on an as-fed basis. The levels of dietary Met were obtained by the inclusion of DL-Met at the expense of Asn and Glu to ensure that all test diets were isonitrogenous. Dietary AA concentrations (Table 2) were determined by Degussa AG (Hanau, Germany) using acid hydrolysis and ion exchange chromatography, with postcolumn derivitization with ninhydrin (Llames and Fontaine, 1994). All dietary indispensable nutrients, except Met and Cys, were provided at a level of at least 120% of their requirement according to the NRC (1998).
Tracer Delivery and Breath Collection
Pigs were offered initially 2 of the half-hourly meals when placed in the oxidation (respiratory) chambers. After equilibration of chamber CO2 contents, the half-hourly feeding regimen began, and background 14CO2 expiration was determined for 2 consecutive 30-min periods. Then the next meal was mixed with tracer, contained a priming dose of 3.5 times the half-hourly dose of 313.4 kBq (SD 35.6) of L-[1-14C]Phe (ARC, St. Louis, MO), plus the first of 8 half-hourly doses of tracer. Expired CO2 was sampled in 30-min intervals at feeding. The remaining 4 meals were given at the end of the oxidation study before the afternoon meal to keep the pigs quiet during shutdown of the oxidation system. The equipment used in this study and the method of processing 14CO2 samples has been described in detail by Bertolo et al. (2005).
Calculations and Statistical Analyses
Oxidation rates were expressed as a percentage of the infused dose, and background 14CO2 enrichment was subtracted from infusion plateau 14CO2 enrichment to obtain final plateau enrichment. Plateaus in oxidation were determined as those collection periods in which a regression of oxidation rate on collection period was not significant (P > 0.1) and the CV during the plateau was less than 10%. Data were tested for outliers using the REG procedure (SAS Inst. Inc., Cary, NC). To determine the Met requirement, DL-Met addition was tested for covariables using the MIXED procedure of SAS, with oxidation rate (% of administered dose) as an independent variable and pig as a random effect. Covariables tested for significance included BW, Phe intake, infusion rate, radioactive background, fraction of airflow collected, and respiration chamber. Covariables were retained in the model at P < 0.05; only pig was retained in the model. Oxidation rates, for all pigs together and individually, were analyzed by 2-phase linear regression to determine minimal Met requirements using the NLIN procedure of SAS. The requirements for individual pigs were averaged, and a SD was calculated. The mean and SD of the Met requirement for the individual pigs were used to calculate the mean Met requirement that would meet the needs of various proportions of the population, using PROC MEANS of SAS and the confidence limits of the MEANS output. The mean requirement estimates determined in the present experiment were compared with requirements calculated according to the NRC (1998), based on the actual mean BW and feed intake of individual pigs, using the Student’s t-test procedure. Significance was taken at P < 0.05.
|
RESULTS
|
|---|
The mean initial and final BW of the pigs were 8.8 kg (SD = 1.5) and 14.0 kg (SD = 2.6), respectively. Mean feed intake was 92.0 g/kg0.75 of BW (SD = 4.7). Two of the 36 observations were removed from data analysis, because no plateau in indicator oxidation was achieved.
Oxidation rates quickly increased until plateaus were obtained. Plateaus in the oxidation rate of L-[1-14C]Phe were achieved quickly and were maintained for 5.2 (SD = 0.8) collection periods. Phenylalanine oxidation was from 9.0 to 21.1% (mean of 15.0%, SD = 3.5) of the infused dose at steady state, depending upon Met intake. The mean of the SE during plateaus in oxidation was 0.5% of the applied dose.
Phenylalanine oxidation was affected (P = 0.001) by dietary Met content and was not affected by any covariates (P > 0.05). However, pig was a significant (P = 0.002) factor in the model. This model resulted in an r2 = 0.75 and was highly significant (P = 0.001). Indicator AA oxidation decreased by 32.0% per each percentage addition (per 10 g/kg) of Met. Indicator AA oxidation (% of dose infused) decreased linearly and in a dose-dependent fashion up to the third level of Met addition (0.29%, P < 0.05) and was not different among the 3 highest levels of Met addition (0.320 to 0.377%, P > 0.20).
The individual requirements of all 6 pigs (Table 3) ranged from 0.320 to 0.373% Met. The mean requirement of all 6 pigs was determined to be 0.340% Met (as-fed basis) with a SD of 0.024 or 7.1% of the mean requirement (Figure 1). The 95% confidence interval for the mean Met requirement was 0.315% (lower CI) and 0.366% (upper CI) Met (Figure 1). The Met content needed to cover 50, 66, and 95% of the population based on the mean requirement plus 0, 1, or 2 SD was 0.340, 0.364, and 0.388%, respectively. Conversely, using confidence limits of the MEANS analysis, we found the upper and lower 95% confidence interval to be 0.315 and 0.366%, respectively (Figure 1).
View larger version (11K):
[in this window]
[in a new window]
|
Figure 1. Indicator AA oxidation response to increasing dietary DL-Met concentration. Breakpoint (requirement) was determined using a 2-phase linear regression model. CI = confidence interval.
|
|
For individual pigs, the Met requirements ranged from 97 to 120% of the estimate according to NRC (1998, Table 3
). The mean estimate of the Met requirement of 0.340% was 104.8% SD 8.6 and was not different from that suggested by NRC (1998) at 0.325% (P = 0.11). The variances of the present estimate of individual Met requirements tended to be greater (P = 0.09) than those according to NRC (1998). The NRC (1998) requirement estimates were correlated (r > 0.94, P < 0.005) with BW and feed intake, but those of the individual estimates from the current experiment were not (r < 0.08, P > 0.88).
|
DISCUSSION
|
|---|
In the current experiment, we determined the interanimal variability of the minimum Met requirement in growing gilts that weighed on average 11.4 kg during the study period. By using the IAAO technique, we were able to test 6 dietary concentrations of Met within approximately 4 wk. By supplying each gilt with each dietary treatment, we were able to determine the Met requirement for each individual pig within the starter to early grower growth stage and to derive an estimate for the requirement plus the associated variability. We chose Met, because it is commonly the second or third dietary limiting AA in young pig diets, and information on the requirement and the associated variability will enable more accurate formulation to maximize lean growth.
To determine the minimum or obligatory Met requirement, Cys (or the dimer of Cys, cystine) needs to be provided at sufficient levels to meet at least its own requirement (Chung and Baker, 1992b; Ball et al., 2006). Cystine can cover the S AA requirement of pigs up to a maximum of 50% (Chung and Baker, 1992a). At dietary concentrations of Cys below its replacement value for Met, some Met is being used to synthesize Cys. Requirements determined under such dietary conditions (Gaines et al., 2005) do not represent the minimum Met requirement. For that reason, we formulated diets with Cys (supplied as L-Cys·HCl) at levels in excess of its maximum replacement value for Met, namely 0.53% as fed, or 75% of the TSAA requirement according to the NRC (1998). Therefore, our results represent the minimum Met requirement.
The present estimate of the population mean total Met requirement (Figure 1, 0.34% of diet) was similar to that recommended by the NRC (1998, 0.325% of diet). The NRC (1998) lists 4 estimates of the minimum Met requirement for pigs below 20 kg of BW. The estimates for the minimum individual pig Met requirement range from 0.29 to 0.48% to 0.52% as fed (Chung and Baker, 1992b; Owen et al., 1995a). Matthews et al. (2001) showed the Met requirement for 10-kg pigs to be 0.39%, whereas Leibholz (1984) suggested 0.25% was sufficient for pigs from 28 to 56 d of age. For pigs of a slightly greater BW (beginning at 19 kg, gaining 625 g/d), Shelton et al. (1951) determined a requirement of 0.3% Met in the presence of adequate Cys, whereas Curtin et al. (1952) estimated the minimum Met requirement as 0.33% for pigs with an initial BW of 13 kg gaining 550 g/d. One of the key factors for differences in requirements is AA availability (Chung and Baker, 1992b). The comparatively high estimate for the Met requirement by Owen et al. (1995b) may have been caused by the use of a Met source of disputed bioavailability (Kim et al., 2006; Yi et al., 2006). The other requirement values fall into a narrow range with a mean of 0.32% (CV 16%). Using tabulated digestibility values (NRC, 1998), we estimated the mean requirement determined in our study as 0.324% Met on a true ileal digestible basis. It is noteworthy that this includes the first 2 published estimates for Met requirement of weaned pigs and indicates that these did not change to a large degree during the past 50 yr.
Common for these requirement estimates is that they were determined under ad libitum feeding, whereas our estimate was derived in pigs fed restrictively. Requirements characterize the point at which a factor other than the tested nutrient becomes limiting for performance. Bertolo et al. (2005) showed that the NRC (1998) program predicts Lys requirements for restrictively fed pigs with good accuracy. Therefore, it is unlikely that Lys was limiting for protein synthesis, because it was provided at 20% above the predicted requirement for restrictively fed pigs (NRC, 1998). Restricting feed intake obviously makes energy intake limiting for weaned pigs; therefore, it is appropriate to express the population mean requirement as a fraction of dietary energy content. Our estimate of 0.23 g of Met/MJ of ME is similar to that by Chung and Baker (1992b) of 0.21 g/MJ of ME and 0.23 g/MJ of ME for baby pigs (Shoveller et al., 2003a), whereas according to Leibholz (1984), Met should be provided at 0.18 g/MJ of ME. The estimate by Leibholz (1984), however, was derived from diets that may have been limiting in Lys and may be an underestimation of the Met requirement. Conversely, supplying AA other than Met at 120% of the requirement (NRC, 1998) as driven by energy intake does not allow an optimal Met-to-Lys ratio from our experiment to be derived.
The present estimate of the mean requirement was 0.34% Met with a CV of 7.0% among animals. The mean Lys requirement for 9 grower pigs (Bertolo et al., 2005) was 0.91% with a CV of 11.7%. The similarity of the interanimal variability of Met and Lys requirements indicates that the variability of requirements may be similar for other AA as well. In this case, a CV of approximately 10% could be an appropriate assumption. The slightly lower variability of mean Met compared with mean Lys requirements may be a consequence of Met requirements not being normally distributed. Compared with the interanimal variability discussed above, statistical variation of requirements previously reported in the literature were derived from experiments that did not allow the estimate of individual requirements. Shoveller et al. (2003a) reported the statistical error of Met requirements as 22% of the estimated requirement in baby pigs when excess dietary Cys was provided; however, a different statistical model was used for determination of the requirement, and this may account for at least part of the greater variability. In the same piglet model, the statistical error of the Met requirement was 8% when there was no Cys provided in the diet (Shoveller et al., 2003b). Kim and Bayley (1983), also using the IAAO technique, reported the statistical error of the requirement as 12% of the mean Met requirement (0.264%) of 3-kg pigs, compared with 15% of the mean requirement in the current study, based on the 95% confidence interval. Therefore, it appears that the interanimal variability of Met requirements is similar or smaller than the statistical error associated with requirement determination.
The NRC (1998) states that both BW and feed intake are correlated with AA requirements, but in this experiment, BW was not a significant covariate for Met requirement because feed was offered restrictively to the pigs. However, BW of the pigs did not vary substantially in the current experiment, and a definite trend to changing requirements only becomes apparent if a greater BW range is compiled. Furthermore, BW was confounded with Phe intake, which was identified as a significant covariate along with portion of airflow collected. Because the indicator AA (Phe) is supplied in excess of the requirement of the pigs, any differences in dietary Phe intake will directly affect indicator oxidation. This effect is independent of the effect of the AA under study. The Phe intake varied due to random differences in analyzed dietary Phe contents and because the feed intake varied among pigs on study days. The significance of the covariate portion of airflow collected indicates that CO2 collection was not quantitative and that it differed among individual oxidation studies. However, when comparing groups, the portion of airflow collected was not a significant factor (P = 0.23), indicating that the estimate of requirement was not affected.
The present experiment used the IAAO to determine Met requirements in weaned pigs. The short adaptation period allowed us to make an estimate of the Met requirement in individual pigs within a narrow weight range (11.3 ± 0.4 kg) and then allowed calculation of the variability of requirements. To account for 95% of the population requirement, Met must be supplied at 0.388% of the diet; however, in most situations, this may not be economically practical. Knowing the variability in the requirement does allow for adjustment of the feeding strategy to optimize herd performance and formulate cost-effective diets.
|
Footnotes
|
|---|
1 This study was funded by Alberta Pork, the Alberta Agricultural Research Institute, and Degussa GmbH. 
2 Current address: Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1.
3 Corresponding author: ron.ball{at}ualberta.ca
Received for publication September 6, 2006.
Accepted for publication October 1, 2007.
|
LITERATURE CITED
|
|---|
Ball, R. O., G. Courtney-Martin, and P. B. Pencharz. 2006. The in vivo sparing of methionine by cysteine in sulfur amino acid requirements in animal models and adult humans. J. Nutr. 136:1682S–1693S[Abstract/Free Full Text]
Bertolo, R. F., S. Moehn, P. B. Pencharz, and R. O. Ball. 2005. Estimate of the variability of the lysine requirement of growing pigs using the indicator amino acid oxidation technique. J. Anim. Sci. 83:2535–2542.[Abstract/Free Full Text]
Brunton, J. A., R. O. Ball, and P. B. Pencharz. 1998. Determination of amino acid requirements by indicator amino acid oxidation: Applications in health and disease. Curr. Opin. Clin. Nutr. Metab. Care 1:449–453.[CrossRef][Medline]
Chung, T. K., and D. H. Baker. 1992a. Maximal portion of the young pig’s sulfur amino acid requirement that can be furnished by cystine. J. Anim. Sci. 70:1182–1187.[Abstract]
Chung, T. K., and D. H. Baker. 1992b. Methionine requirement of pigs between 5 and 20 kilograms body weight. J. Anim. Sci. 70:1857–1863.[Abstract]
Coleman, R. A., R. F. Bertolo, S. Moehn, M. A. Leslie, R. O. Ball, and D. R. Korver. 2003. Lysine requirements of pre-lay broiler breeder pullets: Determination by indicator amino acid oxidation. J. Nutr. 133:2826–2829.[Abstract/Free Full Text]
Curtin, L. V., J. K. Loosli, J. Abraham, H. H. Williams, and L. A. Maynard. 1952. The methionine requirement for the growth of swine. J. Nutr. 48:499–508.[Abstract/Free Full Text]
Gaines, A. M., G. F. Yi, B. W. Ratliff, P. Srichana, D. C. Kendall, G. L. Allee, C. D. Knight, and K. R. Perryman. 2005. Estimation of the ideal ratio of true ileal digestible sulfur amino acids:lysine in 8- to 26-kg nursery pigs. J. Anim. Sci. 83:2527–2534.[Abstract/Free Full Text]
Kim, K. I., and H. S. Bayley. 1983. Amino acid oxidation by young pigs receiving diets with varying levels of sulfur amino acids. Br. J. Nutr. 50:383–390.[CrossRef][Medline]
Kim, B. G., M. D. Lindemann, M. Rademacher, J. J. Brennan, and G. L. Cromwell. 2006. Efficacy of DL-methionine hydroxy analog free acid and DL-methionine as methionine sources for pigs. J. Anim. Sci. 84:104–111.[Abstract/Free Full Text]
Leibholz, J. 1984. Methionine supplementation of diets for pigs between 7 and 56 days of age. Anim. Prod. 39:125–130.
Llames, C. R., and J. Fontaine. 1994. Determination of amino acids in feeds: Collaborative study. J. AOAC Int. 77:1362–1402.
Matthews, J. O., L. L. Southern, and T. D. Bidner. 2001. Estimation of the total sulfur amino acid requirement and the effect of betaine in diets deficient in total sulfur amino acids for the weanling pig. J. Anim. Sci. 79:1557–1565.[Abstract/Free Full Text]
Moehn, S., R. F. P. Bertolo, P. B. Pencharz, and R. O. Ball. 2004. Indicator amino acid oxidation responds rapidly to changes in lysine or protein intake in growing and adult pigs. J. Nutr. 134:836–841.[Abstract/Free Full Text]
Moehn, S., R. F. Bertolo, P. B. Pencharz, and R. O. Ball. 2005. Development of the indicator amino acid oxidation technique to determine the availability of amino acids from dietary protein in pigs. J. Nutr. 135:2866–2870.[Abstract/Free Full Text]
NRC. 1998. Nutrient Requirements for Swine. 10th ed. Natl. Acad. Press, Washington, DC.
Owen, K. Q., R. D. Goodband, J. L. Nelssen, M. D. Tokach, and S. S. Dritz. 1995a. The effect of dietary methionine and its relationship to lysine on growth performance of the segregated early-weaned pig. J. Anim. Sci. 73:3666–3672.[Abstract]
Owen, K. Q., J. L. Nelssen, R. D. Goodband, M. D. Tokach, L. J. Kats, and K. G. Friesen. 1995b. Added dietary methionine in starter pig diets containing spray-dried blood products. J. Anim. Sci. 73:2647–2654.[Abstract]
Pencharz, P. B., and R. O. Ball. 2003. Different approaches to define individual amino acid requirements. Annu. Rev. Nutr. 23:101–116.[CrossRef][Medline]
Shelton, D. C., W. M. Beeson, and E. T. Mertz. 1951. The effect of methionine and cystine on the growth of weanling pigs. J. Anim. Sci. 10:57–64.[Abstract/Free Full Text]
Shoveller, A. K., J. A. Brunton, J. D. House, P. B. Pencharz, and R. O. Ball. 2003a. Dietary cysteine reduces the methionine requirement by an equal proportion in both parenterally and enterally fed piglets. J. Nutr. 133:4215–4224.[Abstract/Free Full Text]
Shoveller, A. K., J. A. Brunton, P. B. Pencharz, and R. O. Ball. 2003b. The methionine requirement is lower in neonatal piglets fed parenterally than in those fed enterally. J. Nutr. 133:1390–1397.[Abstract/Free Full Text]
Yi, G. F., A. M. Gaines, B. W. Ratliff, P. Srichana, G. L. Allee, K. R. Perryman, and C. D. Knight. 2006. Estimation of the true ileal digestible lysine and sulfur amino acid requirement and comparison of the bioefficacy of 2-hydroxy-4-(methylthio)butanoic acid and DL-methionine in eleven- to twenty-six-kilogram nursery pigs. J. Anim. Sci. 84:1709–1721.[Abstract/Free Full Text]
Top
Abstract
INTRODUCTION
