Characterization of feed intake profiles of Axiom suckling sows and link with their performance

Constantly changing prolificity and modified nutritional requirements

In France, the prolificacy of sows has increased considerably in recent decades. This is largely due to the genetic selection of pig lines, with a strong focus on improving productivity. This strategy has resulted in an increase in the number of piglets to be suckled, and therefore in a change in the nutritional requirements of sows during lactation. During this period, the sow’s needs must be covered by her feed intake and/or by mobilizing her body reserves. In the event of deficiency, milk production can be adversely affected, compromising piglet growth and even ^survival2.

While studies have shown that high mobilization of body reserves during lactation can adversely affect sows’ reproductive performance in the following ^cycle8, studies have also shown that the level of feed intake and the trajectory of the feed intake curve have an impact on sows’ maternal performance4^,5,9. In this context, it is worth investigating the diversity of feed intake profiles observed in lactating sows, in order to assess whether there is a link between their feeding behaviour, lactation performance and reproductive performance.

A robust study based on a comprehensive database

This study is based on the longitudinal follow-up, between December 2019 and January 2023, of a herd of 350 sows under Tai Zumu selection. The sows were reared in a seven-band system, with piglets weaned at 28 days. Feed quantities were distributed by a Modulosec® individual feeding system (Skiold, Sæby, Denmark) with daily recording of distributions. This system distributes the ration for one meal in several instalments, allowing automatic modulation of the programmed curve throughout the day. The average net energy content of the feed distributed was 10.0 MJ/kg, and the digestible lysine content was 9.8 g/kg. Daily feed intakes were recorded from maternity entry (ME), on average 7 days before farrowing (MB), to weaning (W), on average 27.4 days after MB, over the course of 2,308 lactations. With an average of 2.65 lactations per sow, 870 distinct sows make up the database for this study.

As the duration of lactation varies according to the day of farrowing, it was considered that a lactation corresponded to a complete period of 24 days following farrowing. Thus, for all sows in the study, the feed intake used for the analysis was from farrowing day (D1) to the ^24th day of lactation (D24).

In addition to daily distributions, the study is based on a database created from data routinely entered on the farm. This database includes

1. individual piglet weights in the first 24 hours of life and at 19 days of age;
2. weights and thicknesses of backfat of sows entering and leaving the maternity ward;
3. numbers of piglets born total (NT), born alive (NV) and weaned (SEV) per litter and for all cycles over the study period;
4. weaning to first mating interval (ISS1, d).

The mobilization of body reserves during lactation was estimated by the relative weight loss (LOSS_PDS, %) of sows between farrowing and weaning. The average daily weight gain of the litter between D1 and D19 (GMQ1-19, kg/d) was calculated, as was the sows’ total feed intake during lactation (CONSO_LACTA, kg).

Distinguishing intake profiles: a novel approach that works

Work on the feed consumption trajectories of lactating sows is still relatively limited, and does not provide a reference method for defining profiles. In this study, we chose to characterize the trajectory of each of the 2,308 feed intake curves via two linear regressions. The first regression was designed to model intake between D1 and D11, identified as the period of greatest increase in quantities distributed. The second linear regression was designed to model intake between D12 and D24, identified as the consumption plateau.

A principal component analysis was then performed on the four variables corresponding to the parameters of the linear regressions previously defined: slopes and intercepts. Next, a hierarchical ascending classification was carried out to distinguish lactations according to the shape of the sows’ feed intake curve, thus creating homogeneous groups in terms of feed intake profiles.

To our knowledge, the method proposed in this study has not been used in other publications. However, the profiles obtained are in line with those described in the literature, as is the number of trajectories identified5^,9. For the purposes of this study, the method used seemed sufficient to distinguish six groups of ingestion curves, with relatively homogeneous numbers per group.

Following these analyses, groups G1 to G6 were identified in ascending order of average CONSO_LACTA: 122.1; 127.5; 133.1; 136.8; 143.8 and 146.6 kg/sow respectively.

Distinct profiles for equivalent performance

Analysis of the sows’ reproductive results showed independence with groups of intake profiles, as well as with criteria associated with feed intake, body reserve mobilization and litter growth. These results are in contradiction with many authors who have highlighted the negative effects of high body reserve mobilization on sow reproductive criteria in the following cycle6^,3. It is possible that these results are linked to the specificity of the sows in this line. In this respect, Banville (2016⁾¹ demonstrated that Tai Zumu sows, unlike European sows, maintained an excellent level of reproductive capacity, regardless of the level of mobilization of their body reserves in the previous lactation.

^1ANOVA– *** : P < 0.001; ** : P < 0.01; * : P < 0.05; ns : P > 0.05

Differentiated strategies to cover nutritional requirements during lactation

Given the negative correlation between CONSO_LACTA and LOSS_PDS, it seems that the lactations making up groups G1 and G2 come from sows that have mobilized their body reserves the most. On the other hand, lactations in group G6 come from sows which have mobilized their body reserves the least. In both cases, milk production and therefore piglet growth are unaffected. Based on the work of Noblet et al. (1988⁾⁷, we can conclude that the sows in these three groups have succeeded in meeting their energy requirements through a system of compensation between ingestion and mobilization of body reserves.

Sows in the G5 profiles, and to a lesser extent those in the G4 profile, combined a high level of feed intake with above-average relative weight loss without any significant increase in milk production, which is problematic. Conversely, in G3 lactations, sows consumed both less feed than average and also mobilized their body reserves less, without impairing milk production.

Conclusion

This study enabled us to identify and describe six distinct groups of suckling sow feed intake profiles, and thus gain a better understanding of the variability of sow requirements. However, the approach used in this article did not reveal any ideal profiles for selection. A further step would be to convert the resources mobilized (ingestion and body reserves) and exported (litter growth) by sows into a single unit of measurement, megajoules of net energy.

Editor(s)

• Article written by Thelma VAN Gheluwe (Project Manager) & Clément Girres (France Manager)
• Article written by the authors and the Axiom Communications Department