Effect of body composition on diet selection in finishing pigs1

Abstract

The objective of this work was to determine if pigs of similar BW, but differing in 10th-rib backfat thickness, would differ in their selection of dietary protein. The first experiment was conducted with gilts and was designed in a 2 x 2 factorial arrangement with the main effects of body fat (lean vs. fat) and feeding program (single vs. choice). Crossbred gilts (n = 32) with an initial BW of 79 kg were sorted into high (2.24 cm, fat) and low backfat (1.51 cm, lean; P < 0.001) groups based on real-time ultrasound scans at the 10th rib. Diets were 1) a low-protein corn-based diet supplemented with crystalline AA (8.5% CP, 0.58% Lys) and 2) a high-protein corn- and soybean meal-based diet (22.7% CP, 1.275% Lys). During wk 1, all pigs were fed a 50:50 mix of the low-protein and high-protein diets. From d 7 to 28, one-half the pigs in the lean and fat groups were given a choice of low- and high-protein diets in separate feeders, whereas the others remained on the 50:50 mix. Initial differences in 10th-rib backfat thickness were maintained through d 28 (fat, 3.02 cm; lean, 2.42 cm; P < 0.001). Circulating leptin concentrations were greater in the fat pigs than in the lean pigs (3.84 vs. 3.35 ng/mL; P < 0.05). Average daily gain (1.06 kg/d) and ADFI (2.64 kg/d) were not different between treatment groups. However, the pattern of selection was different in the choice groups such that lean pigs consumed more (64.4%) of the high-protein diet than did fat pigs (35.6%, P < 0.01), resulting in a difference in the percentage of protein consumed (lean = 16.9% CP; fat = 12.8% CP; P < 0.01). A second experiment of similar design, but with barrows (n = 32; initial BW 69 kg), was conducted. Barrows classified as fat had 2.33 cm of 10th-rib backfat as compared with 1.75 cm in the lean group (P < 0.01). The percentage of protein in the diet selected by the fat barrows was not different from that selected by the lean barrows (15.9 vs. 17.4%). The lack of difference in selection pattern in barrows may be accounted for by the relatively smaller difference in body composition between the fat and lean classifications in the barrows in the choice feeding program as compared with body composition of the gilts. These results demonstrate that gilts of similar BW, but differing in composition, likely have different nutrient requirements and diet selection patterns. Although allowing pigs to self-select did not alter growth rate, the results indicated that diet selection can be used to allow pigs to meet their individual nutrient requirements more closely.