Objective To study the effects of Shenqi-fuzheng injection (SQ) in different doses injected into Zusanli acupoint in treating heart-qi deficiency syndrome and explore its dose-to-effet relationship.Methods The rat model of heart-qi deficiency syndrome was established by forced burden swimming and gavage of large doses of propranlolum. Seven groups of rats were set up with 8 rats in each group: the blank control group, the model group and 5 treatment groups which received a Chinese medicine Shenqi-fuzheng injection (SQ) in different doses (0.05 mL, 0.10 mL, 0.15 mL, 0.20 mL and 0.25 mL per rat, respectively) at ST36 acupoints. The treatment was administered for consecutive 10 days. The general conditions and symptoms of the rats were observed and recorded. The serum concentrations of atrial natriuretic peptide (ANP) and cyclic adenosine monophosphate (cAMP) were assayed by ELISA. The activity of serum superoxide dismutase (SOD) was detected by colorimetry. The pathological changes of myocardial tissue were observed using HE staining. Results Compared with the blank control group, the model group showed obvious heart failure symptoms such as fatigue, purple tongue, short and rapid breath, the concentration of ANP was increased while cAMP decreased, and SOD activity was declined (P<0.001 for all). The pathology of myocardial tissue showed inflammatory cell infiltration and seriously swollen cardiomyocytes in disordered arrangement. Compared with the model group, all treatment groups showed that symptoms were relieved, ANP concentration was decreased, and cAMP concentration and SOD activity were increased. Among them, the changes of indicators in the SQ-0.05 mL group were mildest (P<0.05) while strongest in the SQ-0.20 mL group (P<0.001). The pathological changes of myocardial tissue in the SQ-0.20 mL group were mildest and close to normal. Conclusions ST36 acupoint injection with the Chinese medicine Shenqi-fuzheng injection is an effective therapy for heart-qi deficiency syndrome in rats. The treatment effects are positively correlated with the doses of drug in the range of from 0.05 mL to 0.20 mL per rat.