**Background:** Colostrum is the first food for newborns and provides passive immunity via immunoglobulins. In ruminants, colostrum quality is typically assessed by IgG content using laboratory methods (ELISA, radial immunodiffusion), which require specialized equipment. The Brix refractometer offers a rapid on-farm alternative. This study aimed to evaluate correlation patterns between Brix refractive index and bulk tank colostrum parameters (gross composition, fatty acids, minerals, color) using multivariate approaches in Valle del Belice ewes.
**Methods:** Sixty pooled colostrum samples were collected across two lambing seasons: summer (SSL, end August to early October) and winter (WSL, end December to early February), with 30 samples per season. During SSL, ewes were confined and fed hay ad libitum plus 300 g/head/d barley grain. During WSL, ewes grazed on sulla meadows (Sulla coronaria L.) without hay or concentrates. Samples were analyzed for Brix refractive index (optical refractometer, 0–40% scale, ±0.2% accuracy), gross composition (protein, fat, lactose, urea via infrared method; SCC via flow cytometry), pH, color (CIE LAB: L*, a*, b*), fatty acid profile (GC-FID after bimethylation), and minerals (Ca, P, K, Mg, Na via MP-AES after microwave acid digestion). Statistical methods included Pearson correlation, stepwise discriminant analysis (SDA, forward selection, p=0.15 entry threshold), and multivariate factor analysis (MFA, VARIMAX rotation, eigenvalue >1 criterion).
**Key Results:** Mean Brix was 21.02 ± 1.98%. Gross composition means: protein 12.10 ± 1.73%, fat 8.01 ± 1.51%, lactose 3.13 ± 0.45%, urea 47.94 ± 14.44 mg/dL. SCC log10 was 5.81 (≈645,000 cells/mL), mean pH 6.36 ± 0.18. Minerals (mmol/L): Ca 26.26 ± 6.28, P 82.87 ± 10.99, K 15.11 ± 3.39, Mg 5.06 ± 1.47, Na 13.42 ± 5.52. Fatty acids (g/100g FA): SFA 59.43 ± 3.24, MUFA 31.77 ± 2.50, PUFA 8.80 ± 2.66, ω-3 1.11 ± 0.53, ω-6 3.01 ± 0.45. Color: L* 84.09 ± 2.39, a* −7.87 ± 1.00, b* 19.03 ± 2.81. The strongest Pearson correlation was between Brix and protein (r=0.90, p<0.001). Protein was positively correlated with urea (0.81), fat (0.38), Ca (0.39), Mg (0.58), MUFA (0.34), yellowness (0.78), and negatively with lactose (−0.41), SFA (−0.32), lightness (−0.45), redness (−0.37). SCC and pH were positively correlated (0.49). SCC was positively correlated with Na (0.37) and negatively with K (−0.28). SDA showed protein alone explained 80.54% of Brix variability (R²=0.8054). Adding Na, Mg, Ca, lactose, b*, ω-6, and K increased R² to 0.9083. MFA extracted three factors: Factor 1 (68.4% variance, "Colostrum quality of grazing sheep") correlated with SCC, pH, PUFA, ω-3; Factor 2 (23.3%, "Good quality colostrum") correlated with Brix, protein, fat, urea, P, MUFA, yellowness; Factor 3 (8.3%, "Mineral component") correlated with Ca, K, Mg, Na.
**Clinical Implications:** The Brix refractometer is confirmed as a practical, rapid on-farm tool for estimating sheep colostrum nutritional quality, given its strong correlation with protein content (r=0.90). The multivariate approach identified that protein percentage, calcium, and magnesium together explained over 85% of Brix variability. Understanding correlations between colostrum parameters—particularly the relationships between SCC, pH, and mineral shifts (Na increase, K decrease) as indicators of subclinical mastitis—enables farmers to assess udder health and colostrum quality. The three latent factors provide a framework linking colostrum composition to management practices (grazing vs confinement) and nutritional quality. These findings support targeted interventions for optimal lamb nutrition in the first days of life.