Precise identification of deficient intersegmental coordination patterns and functional limitations is conducive to the evaluation of surgical outcomes after total knee arthroplasty (TKA) and the design of optimal personalized rehabilitation protocols. However, it is still not clear how and when intersegmental coordination patterns change during walking, and what functional limitations are in patients with TKA. This study was designed to investigate lower limb intersegmental coordination patterns in patients with knee osteoarthritis before and after TKA and identify how intersegmental coordination of patients is altered during walking before and after TKA. It was hypothesized that 6-month after TKA, intersegmental coordination patterns of patients are improved compared with that before TKA, but still do not recover to the level of healthy subjects. Gait analysis was performed on 36 patients before and 6-month after TKA and on 34 healthy subjects. Continuous relative phase (CRP) derived from the angle-velocity phase portrait was used to measure the coordination between interacting segments throughout the gait cycle. Thigh-shank CRP and shank-foot CRP were calculated for each subject. Statistical parametric mapping (SPM), a one-dimensional analysis of the entire gait cycle curve, was performed directly to determine which periods of the gait cycle were different in patients and healthy subjects. Six-month after TKA, thigh-shank CRP was significantly higher during 5–12% of the gait cycle (p = 0.041) and lower during 44–95% of the gait cycle (p < 0.001) compared with healthy subjects, and was significantly higher during 62–91% of the gait cycle (p = 0.002) compared with pre-operation. Shank-foot CRP was significantly lower during 0–28% of the gait cycle (p < 0.001) and higher during 58–94% of the gait cycle (p < 0.001) compared with healthy subjects, and was significantly lower during 3–18% of the gait cycle (p = 0.005) compared with pre-operation. This study found that patients exhibited altered intersegmental coordination during the loading response and swing phase both before and after TKA. Six-month after TKA, the thigh-shank coordination was partially improved compared with pre-operation, but still did not recover to the level of healthy subjects, while there was no improvement in the shank-foot coordination pattern after TKA compared with pre-operation. CRP combined with SPM methods can provide insights into the evaluation of surgical outcomes and the design of rehabilitation strategy.