Efficacy and Safety of Manipulative Therapy for Frozen Shoulder: a Meta-analysis

Abstract

Background In recent years, there have been many randomized controlled trials (RCTs) featuring manipulation for frozen shoulder, but there is a lack of new systematic reviews on the efficacy and safety of manipulation for frozen shoulder. Objective To evaluate the efficacy and safety of manipulation in the treatment of frozen shoulder by Meta-analysis, and understand whether "rotation shoulder joint" could affect the effect of manipulative therapy by subgroup. Methods PubMed, Cochrane Library, Embase, Medline, CNKI, China Biomedical Literature Database, Wanfang Data were searched for RCTs on the efficacy and safety of manipulative therapy for frozen shoulder, with manipulative therapy as the intervention of experimental group and non-pharmacological therapy as the intervention of control group from inception to 2023-03-01. Two investigators extracted data and evaluated the quality. Meta-analysis was performed using Revman 5.3 software. Results A total of 12 RCTs were included in this Meta-analysis, with a total of 893 patients with frozen shoulder, including 451 patients in the experimental group and 442 cases in the control group. Results of Meta-analysis showed that the improvement of VAS by manipulative therapy was better than physical therapy, acupuncture therapy and conventional therapy (SMD=1.09, 95%CI=0.81-1.37, P<0.000 01; SMD=1.05, 95%CI=0.31-1.79, P=0.006; SMD=0.96, 95%CI=0.67-1.26, P<0.000 01) ; subgroup analysis of manipulative therapy compared with physical therapy showed a significant difference in effect between those with and without "rotation shoulder joint" (Z=4.39, P=0.04). The improvement of Constant-Murley score in manipulative therapy was better than that in physical therapy (MD=2.79, 95%CI=2.27-3.32, Z=10.41, P<0.000 1). The improvements of passive flexion (SMD=1.40, 95%CI=0.10-2.70, P=0.03), passive abduction (SMD=1.45, 95%CI =0.18-2.71, P=0.02) and passive external rotation (SMD=1.77, 95%CI=0.18-3.36, P=0.03) by manipulative therapy were better than those in the physical therapy group; subgroup analysis showed a significant difference between the effects of the manipulative therapy with and without "rotation shoulder joint" in improving passive flexion and passive abduction (Z=7.34, P=0.007; Z=2.25, P=0.03). The overall effective rate of manipulative therapy was higher than that of physical therapy (RR=1.16, 95%CI=1.02-1.32, P=0.03) ; there was no statistically significant difference in the overall effective rate between the manipulative therapy and acupuncture therapy (RR=1.24, 95%CI=1.00-1.54, P=0.05). The cure rate of manipulative therapy was better than that of physical therapy group (RR=3.71, 95%CI=1.29-10.67, P=0.01) and acupuncture group (RR=1.79, 95%CI=1.09-2.94, P=0.02). Egger's test and Begger's test suggested no significant publication bias (P=0.66, 0.66). Cases of fractures and other traumas caused by manipulative therapy were not reported in the included studies. Conclusion The existing literature evidence shows that manipulation can effectively reduce the pain and shoulder joint dysfunction in patients with frozen shoulder, improve the mobility of shoulder joint, overall effective rate and the cure rate of patients, the manipulative therapy with "rotation shoulder joint" can further improve the efficacy in terms of VAS improvement, passive flexion and passive abduction, the overall safety of manipulative therapy is relatively high, but more large samples and high-quality RCTs are still needed to further confirm the results.

Keywords

• frozen shoulder|bursitis|manipulation|shoulder rotation|randomized controlled trial|meta-analysis