OBJECTIVE: To evaluate the efficacy of continuous femoral block on the postoperative analgesia and functional recovery after total knee arthroplasty (TKA). METHODS: Two hundreds and eighty patients who underwent TKA were randomized into two groups:the group receiving continuous femoral block (CFNB) and the group receiving patient controlled intravenous analgesia (PCIA), each group included 140 participants. Femoral nerve block with ropivacaine by ultrasonic guidance was performed in group CFNB and group PCIA were administrated with patient controlled intravenous analgesia. Numerical rating scale (NRS) scores at rest and in motion at 24, 48, 72 h, 3, 6 and 12 months postoperatively, also the NRS scores at hospital discharge were recorded. The incidence of moderate-severity pain, as well as the degree of knee flexion and the WOMAC scores at 3, 6 and 12 months after surgery were analyzed. The rescue analgesic administration and analgesia-related adverse effects were also recorded. Data were expressed as mean± standard deviation (SD) for normally distributed continuous variables and total number (percent frequency) for categorical variables. If non-normally distributed, data were expressed median inter-quartile range. Student's t-test, Wilcoxon rank test were used to compare results for continuous variables, when appropriate. Chi-square test was used to compare results for categorical variable, Fisher exact test was used for categorical variables when the number of event was less than 5. RESULTS: NRS scores of group CFNB in motion was 3 (3-4) at discharge time, and 3 (2-4), 3 (2-3) at 3 months and 6 months postoperatively, while the scores of group PCIA was 4 (4-4), 3 (3-4), 3 (3-4), respectively. And at rest, NRS scores of group CFNB was 3 (2-3), 1 (1-2), 1 (1-1) at discharge time, and 3, 6 months postoperatively. Compared with group PCIA, NRS scores in motion of group CFNB at discharge time (Z=-5.174, P<0.05) and 3 months (Z=2.308, P=0.021), as well as 6 months postoperatively (Z=-2.495, P=0.013), were significantly lower,also for the NRS scores at rest (Z=-2.405, P=0.016; Z=-4.360, P<0.05; Z=-9.268, P<0.05). The degree of knee flexion of group CFNB at 3 and 6 months postoperatively was 92 (88-97), 103 (99-106), while the degree of knee flexion of group PCIA was 89 (86-95), 100 (97-105); the WOMAC scores of group CFNB at 3 and 6 months postoperatively was 21 (18-26), 18 (16-22), while the scores of group PCIA was 24 (20-27), 21 (17-24). WOMAC scores of group CFNB was lower compared with group PCIA at 3 (Z=-2.467, P=0.014) and 6 (Z=-2.537, P=0.011) months postoperatively while the degree of knee flexion of group CFNB was higher (Z=-2.175, P=0.030; Z=-2.471, P=0.013). Moreover, the frequency of bolus and frequency of rescue of group CFNB was 2.3 and 0.6, while the frequency of group PCIA was 2.6 and 1.1, the frequency of bolus and frequency of rescue were lower in group CFNB (t=-2.984, P=0.003; t=-3.213, P=0.002). The incidence of adverse events such muscle weakness of low limbs,nausea and vomiting were similar in two groups (P>0.05). CONCLUSION: CFNB can alleviate the postoperative pain after TKA with safety, help improving the short-middle-term functions of knee and quality of patients' lives.