Physiological proxies of feed efficiency in pigs: exploring the predictive value of blood biochemical and hematological profiles
Frontiers in Veterinary Science · 05.11.2026 · 13
DOI: 10.3389/fvets.2026.1734232 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Veterinary Science, online erschienen am 05.11.2026. Verfasst von Aitor Arrazola und 3 weiteren Autor:innen. Beteiligte Einrichtung: Animal Breeding and Genetics Program, IRTA, Torre Marimon, Caldes de Montbui, Spain. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Developing strategies to improve feed efficiency can help the swine industry become more sustainable while meeting the worldwide demand for pork products, yet most commercial barns cannot practically calculate direct metrics of individual feed efficiency. Physiological indicators of metabolic pathways can serve as intermediate variables to predict performance outcomes and, therefore, as biomarkers of feed efficiency and other desirable production traits in growing pigs. This research aimed to assess the role of conventional blood parameters as proxies for performance outcomes and their predictive value using multivariate analyses and machine learning techniques. We considered data from 176 pigs (out of 762 pigs) across 5 trials. Pigs were reared in groups of 12–14 pigs per pen under similar conditions from 80 to 165 days old. Feed intake, body weight, and backfat thickness were recorded periodically during rearing, and two blood samples were collected from three pigs per pen for biochemical and hematological analyses at 125 and 165 days old. The association between blood parameters and performance outcomes was described using correlation and multiple regression analyses, and the predictive value of the metabolic and hematological profile was assessed by cross-validation after using machine learning techniques to obtain a regularized regression model using an Elastic Net. Among the blood parameters considered, serum concentrations of urea, cholesterol, albumin, globulins, ALP, and the mean corpuscular volume of erythrocytes stood out as potential biomarkers of production and, production efficiency traits during rearing. Regression regularization further supported the importance of these biological traits, along with creatinine and bilirubin concentrations, as partial predictors of feed efficiency and average daily performance outcomes. All these blood parameters create a physiological signature of value to forecast performance phenotypes in healthy growing pigs, capable of predicting up to a fourth of the variability of feed efficiency and one fifth of the variability of average daily feed intake and growth rate. These results highlight that integrating conventional information on physiological metabolites related to protein metabolism (from proteolysis and muscle degradation), nutritional state, and liver function can further explain and predict production and, production efficiency traits in healthy growing pigs.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Animal Breeding and Genetics Program, IRTA, Torre Marimon, Caldes de Montbui, Spain.
Zitieren
APA
Arrazola, A., Piñero, M., Sánchez, J. P., & Quintanilla, R. (2026). Physiological proxies of feed efficiency in pigs: exploring the predictive value of blood biochemical and hematological profiles. Frontiers in Veterinary Science, 13. https://doi.org/10.3389/fvets.2026.1734232
Vancouver
Arrazola A, Piñero M, Sánchez JP, Quintanilla R. Physiological proxies of feed efficiency in pigs: exploring the predictive value of blood biochemical and hematological profiles. Frontiers in Veterinary Science. 2026;13. doi:10.3389/fvets.2026.1734232
BibTeX
@article{Arrazola2026,
author = {Arrazola, Aitor and Piñero, M and Sánchez, J P and Quintanilla, Raquel},
title = {Physiological proxies of feed efficiency in pigs: exploring the predictive value of blood biochemical and hematological profiles},
journal = {Frontiers in Veterinary Science},
year = {2026},
volume = {13},
doi = {10.3389/fvets.2026.1734232}
}
EUVIMED