ObjectiveTo fabricate an injectable composite bone substitute with hyaluronic acid (HA) and calcium sulfate and to evaluate the biocompatibility and effect of the composite on cell proliferation, osteogenic differentiation in vitro and osteogenic capability in vivo. MethodsCalcium sulfate powder was mixed with HA solution, cross-linked HA solution, and phosphate buffer solution (PBS) in a ratio of 2∶1 (W/V) to get composites of CA+HA, CA+HAC, and CA. The standard extracts from above 3 materials were prepared according to ISO10993-5, and were used to culture mouse MC3T3-E1 cells. The composite biocompatibility and cell proliferation in different concentrations of extract were tested with cell counting kit-8 (CCK-8). The cells were cultured with standard medium as a control. The optimal concentration was selected for osteogenic differentiation test, and ELISA Kit was used to determine the alkaline phosphatase (ALP), collagen type I (COL-I), and osteocalcin (OCN). The femoral condylar bone defect was made on New Zealand white rabbits and repaired with CA+HA, CA+HAC, and CA. Micro-CT was done to evaluate new bone formation with bone volume/tissue volume (BV/TV) ratio at 6 and 12 weeks. HE staining was used to observe bone formation. ResultsCA+HA and CA+HAC were better in injectability and stability in PBS than CA. The biocompatibility test showed that absorbance (A) value of CA group was significantly lower than that of control group (P<0.05) at 6, 12, and 24 hours after culture, but no significant difference was found inA values between CA+HA group or CA+HAC group and control group (P>0.05). The proliferation test showed 25% and 50% extract of all 3 materials had significantly higherA value than control group (P<0.05). For 75% and 100% extract, only CA+HA group had significantly higherA value than control group (P<0.05). And 50% extract was selected for osteogenic differentiation test. At 14 and 21 days, ALP, COL-I and OCN concentrations of CA+HA group and CA+HAC group were significantly higher than those of CA group and control group (P<0.05). Micro-CT results showed higher BV/TV in CA+HA group and CA+HAC group than CA group at 6 and 12 weeks (P<0.05), but no significant difference was found between CA+HA group and CA+HAC group (P>0.05). HE staining revealed that a little bone tissue was seen in CA+HA group and CA+HAC group, but there was no bone formation in CA group at 6 weeks; more streak bone tissue in CA+HA group and CA+HAC group than CA group at 12 weeks. ConclusionComposites prepared with calcium sulfate and HA or with cross-linked HA are stable, injectable, and biocompatible. The materials have excellent effect on proliferation and differentiation of mouse MC3T3-E1 cells. They also show good osteogenic capability in vivo. So it is a potential bone substitutes for bone defective diseases.