Since 2001, purebred Yorkshire pigs at Iowa State University have been divergently selected for increased (Low RFI) and decreased (High RFI) feed efficiency based on single trait selection for residual feed intake (RFI). Using data from generations 2–5 (G2–5), Bunter et al. (2010) found that serum insulin-like growth factor-I (IGF-I) concentration in young pigs (33–42 d) was lower in the Low RFI line, and had a high genetic correlation (0.63) with RFI, suggesting that it could be used as an indicator trait for RFI. The objective here was to validate these findings in G10 of the RFI lines. IGF-I concentration was measured by Rivalea Ltd. (Australia) using the Primegro assay on blood samples collected from 35 to 42 d of age on 377 piglets from G10. IGF-I data were analyzed in ASREML with a univariate sire model and a bivariate animal model with RFI data on 2,370 pigs from G1–10. Both models for IGF-I included the fixed effects of line, sex and contemporary group, covariates of age at blood collection and interval from weaning to blood collection, and random effects of litter, and sire (no pedigree) or animal (with complete pedigree). The concentration of IGF-I (ng/mL) was estimated to be 55.7 (±19.1) and 53.4 (±44.2) lower in the Low RFI line compared with the High RFI line based on the sire (P = 0.01) and animal (P = 0.3) models, respectively. Bunter et al. (2010) reported a similar line difference of 47.0 ng/mL in G5. The estimated genetic correlation between RFI and IGF-I was 0.68 (±0.38), similar to Bunter et al. (2010). In conclusion, lower juvenile IGF-I concentration was validated to be associated with pigs selected for increased feed efficiency based on RFI and, therefore, may serve as an early blood indicator to select for feed efficiency. However, when considering the full pedigree, the line difference was not significant. Funding provided by AFRI-NIFA grant #2011–68004–30336 and by Rivalea Ltd. (Australia).