Prediction of toughness and other beef quality characteristics of the m. longissimus thoracis using polarized near-infrared reflectance spectroscopy

Abstract

Near-infrared (NIR) spectroscopy using polarized light was investigated for its ability to predict Warner Bratzler shear force (WBSF), lightness (L*), redness (a*), and yellowness (b*), drip loss, cooking loss, pH and intramuscular fat of bovine longissimus thoracis (LT, rib eye) muscle at 3 and 12 days post mortem. Muscles were derived from a 2 × 2 × 2 factorial study investigating the effects of residual feed intake (RFI) (low versus high, n = 27 and 21, low and high RFI, respectively), hormonal growth promotants (control versus implanted, n = 24), β-adrenergic agonist use (control versus ractopamine hydrochloride, 200 mg/head/day, last 28 days of finishing, n = 24) on the LT meat quality of crossbred Angus steers at 3 and 12 days post mortem. Near-infrared spectra were collected at days 3 and 12 post-mortem without (N) and with a polarizer (1100 to 1650 nm) with its axis perpendicular (R) and parallel to the muscle fiber direction (P). Analysis of variance indicated that LT from implanted steers had a greater mean WBSF value at day 3 (57.04 ± 1.82 N) than non-implanted steers (45.14 ± 1.78 N), but not at day 12, and that post mortem aging increased L*, a* and b* values (p < 0.01). Partial least square regression indicated that NIR spectra collected in R mode at day 3 gave the best validation r2 for day 3 WBSF (0.81 ± 0.01), drip loss (0.74 ± 0.02) and cooking loss (0.85 ± 0.008). Also for day 3, L* (r2 = 0.76 ± 0.02) and a* (r2 = 0.83 ± 0.01) were best predicted in P mode, while b* was predicted with more than 80% accuracy regardless of polarization. Intramuscular fat (r2 = 0.81 ± 0.01) and day 3 pH (r2 = 0.79 ± 0.03) were best predicted with polarization regardless of axis direction. At day 12 post-mortem, L* was best predicted in P mode (r2 = 0.82 ± 0.006), a* was best predicted in R mode (r2 = 0.72 ± 0.03), b* was best predicted in N and P mode (r2 = 0.80 ± 0.01), WBSF, drip loss, cooking loss, and pH (r2 = 0.76 ± 0.01, 0.73 ± 0.01, 0.73 ± 0.02, and 0.84 ± 0.02, respectively) were best predicted in R mode, and IMF was best predicted in P mode (r2 = 0.78 ± 0.01). Results indicated that polarized NIR spectroscopy is a promising non-invasive method to predict meat quality characteristics.