This experiment was conducted to study the effect of different rearing systems on growth, nutrient utilization and serum indices of early weaned Hu twin lambs. Twenty-four pairs of newborn Hu male twin lambs were equally divided into 2 groups according to paired experiment design. In each pair, one lamb was artificially reared (AR) with milk replacer, while the other one was ewe reared (ER). Body weight was recorded and blood samples were collected every 10 d. Two digestion trials were undertaken during d 45 to 55 and d 75 to 85 with 4 pairs of lambs from each group. On d 60 and 90, another 3 pairs of lambs from each group at each time point were euthanized for weighing gastrointestinal organs. The data of body weight and serum indices (total protein, cholesterol, glucose, triglyceride, albumin, blood urea nitrogen, alkaline phosphatase activity, and immunoglobulin) were analyzed using PROC ANOVA with repeated measure, while the others were analyzed using PROC paired t-test of SAS 9.1. The results showed that the BW of lambs in ER group was higher than that in AR group on d 30 (P < 0.05). No difference was observed in the final BW of AR group and ER group (21.02 vs 21.92 kg, P > 0.05) or the overall average daily gain (205.97 vs 217.01 g/d, P > 0.05). On d 60, the proportion of the rumen to 4 stomaches (weight basis) of lambs in AR group was higher (P < 0.05) while that of omasum and abomasum were lower (P < 0.05) than those in ER group. The proportion of small intestine was higher in AR group than that in ER group (P < 0.05). There was no difference in the nutrient digestibility (P > 0.05) or the serum concentration of total protein, cholesterol, glucose, triglyceride, albumin, blood urea nitrogen, alkaline phosphatase activity, and immunoglobulin (G, A, and M) (P > 0.05). In general, lambs reared with milk replacer achieved the same growth performance as their ewe-reared counterparts.