The short-term effects of dry cupping the lumbar paraspinal muscles in individuals with non-specific low back pain: a single-blind randomized trial

Abstract

Introduction: Non-specific low back pain (NSLBP) is a leading cause of global disability, affecting millions of individuals and imposing significant personal and societal burdens. Current treatments (chiropractic or physiotherapy care, stretching exercises, maintaining a physically active lifestyle, and pharmacological interventions) often yield inconsistent results, highlighting the need for alternative therapies. Dry cupping, an ancient practice involving suction on the skin, has shown promise in improving pain and mobility, yet comprehensive evidence for its efficacy in NSLBP remains limited. This study investigated the short-term effects of dry cupping on lumbar paraspinal muscles in individuals with NSLBP, focusing on pain, range of motion (ROM), skin temperature (as a blood flow indicator), function, and perceived treatment effect. Methods: Adults aged 18–55 (31 females and 21 males) with clinician-confirmed NSLBP were recruited to participate in a single-blind randomized clinical trial. Fifty-two participants (26 intervention, 26 placebo) received three sessions of either dry cupping or placebo cupping with approximately 48 hours in between each session. The intervention group received static cupping (10 minutes/session) at lumbar landmarks, based on indicated painful areas, using Hansol© cups. The placebo group received identical-appearing cups without suction. Measurements included ROM (Sit and Reach Test, dual inclinometry), pressure pain threshold (algometry), subjective pain (Numeric Pain Rating Scale; NPRS), skin temperature (laser thermometer), function (Roland-Morris Questionnaire), and perceived treatment effect (Patient Global Impression of Change Scale; PGICS). Data were analyzed using three-way and two-way mixed factorial ANOVAs and post-hoc tests. Results: A three-way mixed factorial ANOVA revealed a nonsignificant interaction effect between the three variables (treatment group, session and time) F(1.720, 26)=1.172, p=.309, 𝜂2=.023) but a significant two-way interaction effect between treatment group and session (F(1.8949, 26)=29.603, p<.001, η²=.372) and between treatment group and time (F(1, 26)=126.968, p<.001, η²=.717) on Sit and Reach scores. A three-way mixed factorial ANOVA revealed a nonsignificant interaction effect between the three variables (treatment group, session and time) (F(1.766, 26)=.206, p=.787, 𝜂²=.004), but a significant two-way interaction effect between treatment group and session (F(1.558, 26)=25.207, p<.001, η²=.335) and between treatment group and time (F(1, 26)=131.725, p<.001, η²=.725) on inclinometry scores. A three-way mixed factorial ANOVA revealed a nonsignificant interaction effect between the three variables (treatment group, session and time) (F(2, 26)=.361, p=.698, 𝜂²=.007), but a significant two-way interaction effect between treatment group and session (F(2, 26)=13.694, p<.001, η²=.215) and between treatment group and time (F(1, 26)=71.237, p<.001, η²=.604) on pressure pain threshold. A three-way mixed factorial ANOVA revealed a nonsignificant interaction between the three variables (treatment group, session and time) (F(1.593, 26)=.956, p=.371, 𝜂²=.019), but a significant two-way interaction effect between treatment group and session (F(1.545, 26)=11.640, p<.001, η² =.189) and between treatment group and time (F(1, 26)=97.051, p<.001, η²=.660) on subjective pain perception. A three-way mixed factorial ANOVA revealed a significant three-way interaction effect between treatment group, time, and session (F(2, 26)=19.666, p<.001, η²=.282) on skin temperature. A two-way mixed factorial ANOVA revealed a significant interaction effect between treatment group and session (F(1.702, 26)=9.387, p<.001, η²=.158) on perceived treatment effect. A three-way mixed factorial ANOVA revealed a nonsignificant interaction effect between the three variables (treatment group, session and time) (F(2, 26)= 1.538, p=.220, 𝜂²=.030), but a significant two-way interaction effect between treatment group and session (F(2, 26)=6.522, p=.002, η²=.115), between treatment group and time (F(1, 26)=28.736, p<.001, η²=.365) and between session and time (F(2, 26)=4.363, p=.015, 𝜂²=.080) on overall function. Conclusion: This randomized clinical trial demonstrates that dry cupping therapy significantly improves short-term pain and ROM in individuals with NSLBP, with a modest effect on overall function. Compared to placebo, the intervention group showed clinically meaningful reductions in pain, increased ROM, elevated skin temperature, and significantly higher treatment satisfaction, leading to a greater understanding of the physiological mechanisms of treatment. Results highlight the importance of multiple treatment sessions for cumulative effects and demonstrate dry cupping is a viable non-pharmacological adjunct for NSLBP management. Further research is needed to explore the long-term sustainability of these effects and to establish optimal treatment protocols, including frequency and duration.