Objective:To observe the clinical effectiveness of manual therapy based on posture decoding for patients with lower crossed syndrome (LCS).Methods:Thirty-six LCS patients were randomly divided into an observation group and a control group, each of 18. The observation group received manual therapy based on posture decoding, while the control group was treated with proprioceptive neuromuscular facilitation (PNF), both in 20min sessions, once a week for 4 weeks. Before the experiment, after one, two and four weeks of treatment and followed-up 4 and 8 weeks later, both groups were evaluated using a visual analogue scale (VAS), the Oswestry Disability Index (ODI) and finger-floor distance (FFD). Anterior pelvic tilt angles (ASIS-PSISs), sacral slopes (SS), lumbar curve index (LCI) and surface EMG flexion-relaxation ratios (FRRs) were also recorded from both groups before and after the treatment.Results:After one and four weeks of the treatment, the average VAS, ODI, and FFD had decreased significantly in both groups, with all significantly lower in the observation group, on average. At the final follow-up, the average VAS and ODI scores of both groups were significantly lower than before the treatment, with those of the observation group significantly lower than the control group′s averages. After 4 weeks of treatment significant differences were observed also in the group′s average ASIS-PSISs, SSs and LCIs compared with before the treatment. And right after the treatment the left and right surface electromyography FRRs of the observation group were significantly higher than those of the control group.Conclusion:Manual therapy based on posture decoding can significantly improve the pelvis forward angle and lumbar motion of LCS patients, relieving back pain and relaxing back muscles.