Environmental enrichment is a recognized strategy to improve sow welfare. Providing manipulable wood materials and straw pellets to sows during gestation might participate in satisfying their investigative behavior. Sows may also get a benefit from chewing and eating fibers, which can decrease their hunger. Such strategy, however, should not adversely affect metabolic status of sows and newborn piglets. In this study, gestating sows were group-housed in a conventional system on slatted floor (C, n = 26) or in the same system enriched with pieces of oak attached to a chain and straw pellets (CE, n = 30). Straw pellets were provided in the trough at a rate of 200 g/d from 3 to 30 days of gestation (DG) and 400 g/d from 31 to 104 DG. On DG105, sows were transferred into farrowing pens and housed in identical individual stalls on slatted floor. Sow body weight and backfat thickness were recorded before insemination, on DG105 and at weaning. Plasma concentrations of metabolites and insulin were measured in sows on DG73 and DG102. Blood was also collected from a subset of piglets at birth (n = 6 per litter in 5 C and 8 CE litters) to determine their metabolic and oxidative status. Sow body weight, backfat thickness, and estimated body lipid and protein masses did not differ between treatments during gestation and lactation. Maternal plasma concentrations of glucose, lactate, NEFA, beta‑hydroxy‑butyrate and insulin did not differ between groups (P > 0.10). However, CE sows had lower circulating concentrations of urea (P < 0.001) and a greater increase in triglyceride concentrations between DG73 and DG102 (P < 0.05). Treatment did not influence (P > 0.10) piglet birth weight or plasma concentrations of glucose, fructose, lactate and albumin, antioxidant potential and the oxidative stress index in newborns. On the day after farrowing, the milk produced by CE sows contained more minerals (P < 0.01) than the milk of C sows but similar contents of protein, fat, lactose, energy and IgA. The proportion of piglets that died at birth and within 12 h of birth was lower in CE than in C litters (P < 0.05), but overall mortality (stillbirth + preweaning death) did not differ between the two groups. To conclude, providing manipulable materials and straw pellets to sows during gestation had little impact on sow metabolic status but improved early neonatal survival.