ObjectivesThis study aims to investigate the effectiveness of cupping therapy on low back pain (LBP). MethodsMedline, Embase, Scopus and WANFANG databases were searched for relevant cupping RCTs on low back pain articles up to 2023. A complementary search was manually made on 27 September for update screening. Full-text English and Chinese articles on all ethnic adults with LBP of cupping management were included in this study. Studies looking at acute low back pain only were excluded. Two independent reviewers screened and extracted data, with any disagreement resolved through consensus by a third reviewer. The methodological quality of the included studies was evaluated independently by two reviewers using an adapted tool. Change-from-baseline outcomes were treated as continuous variables and calculated according to the Cochrane Handbook. Data were extracted and pooled into the meta-analysis by Review Manager software (version 5.4, Nordic Cochrane Centre). ResultsEleven trials involving 921 participants were included. Five studies were assessed as being at low risk of bias, and six studies were of acceptable quality. High-quality evidence demonstrated cupping significantly improves pain at 2–8 weeks endpoint intervention (d=1.09, 95% CI: [0.35–1.83], p = 0.004). There was no continuous pain improvement observed at one month (d=0.11, 95% CI: [−1.02–1.23], p = 0.85) and 3–6 months (d=0.39, 95% CI: [−0.09–0.87], p = 0.11). Dry cupping did not improve pain (d=1.06, 95% CI: [−0.34, 2.45], p = 0.14) compared with wet cupping (d=1.5, 95% CI: [0.39–2.6], p = 0.008) at the endpoint intervention. There was no evidence indicating the association between pain reduction and different types of cupping (p = 0.2). Moderate- to low-quality evidence showed that cupping did not reduce chronic low back pain (d=0.74, 95% CI: [−0.67–2.15], p = 0.30) and non-specific chronic low back pain (d=0.27, 95% CI: [−1.69–2.24], p = 0.78) at the endpoint intervention. Cupping on acupoints showed a significant improvement in pain (d=1.29, 95% CI: [0.63–1.94], p < 0.01) compared with the lower back area (d=0.35, 95% CI: [−0.29–0.99], p = 0.29). A potential association between pain reduction and different cupping locations (p = 0.05) was found. Meta-analysis showed a significant effect on pain improvement compared to medication therapy (n = 8; d=1.8 [95% CI: 1.22 – 2.39], p < 0.001) and usual care (n = 5; d=1.07 [95% CI: 0.21- 1.93], p = 0.01). Two studies demonstrated that cupping significantly mediated sensory and emotional pain immediately, after 24 h, and 2 weeks post-intervention (d= 5.49, 95% CI [4.13–6.84], p < 0.001). Moderate evidence suggested that cupping improved disability at the 1–6 months follow-up (d=0.67, 95% CI: [0.06–1.28], p = 0.03). There was no immediate effect observed at the 2–8 weeks endpoint (d=0.40, 95% CI: [−0.51–1.30], p = 0.39). A high degree of heterogeneity was noted in the subgroup analysis (I2 >50%). ConclusionHigh- to moderate-quality evidence indicates that cupping significantly improves pain and disability. The effectiveness of cupping for LBP varies based on treatment durations, cupping types, treatment locations, and LBP classifications. Cupping demonstrated a superior and sustained effect on pain reduction compared with medication and usual care. The notable heterogeneity among studies raises concerns about the certainty of these findings. Further research should be designed with a standardized cupping manipulation that specifies treatment sessions, frequency, cupping types, and treatment locations. The actual therapeutic effects of cupping could be confirmed by using objective pain assessments. Studies with at least six- to twelve-month follow-ups are needed to investigate the long-term efficacy of cupping in managing LBP. Trial registrationThis systematic review was initially registered on PROSPERO with registration code: CRD42021271245 on 08 September 2021.