ObjectiveConstraint-induced movement therapy (CIMT) is a common treatment for upper extremity motor dysfunction after a stroke. However, whether it can effectively improve lower extremity motor function in stroke patients remains controversial. This systematic review comprehensively studies the current evidence and evaluates the effectiveness of CIMT in the treatment of post-stroke lower extremity motor dysfunction.MethodsWe comprehensively searched randomized controlled trials related to this study in eight electronic databases (PubMed, Embase, The Cochrane Library, Web of Science, CBM, CNKI, WAN FANG, and VIP). We evaluated CIMT effectiveness against post-stroke lower extremity motor dysfunction based on the mean difference and corresponding 95% confidence interval (95% CI). We assessed methodological quality based on the Cochrane Bias Risk Assessment Tool. After extracting the general information, mean, and standard deviation of the included studies, we conducted a meta-analysis using RevMan 5.3 and Stata 16.0. The primary indicator was the Fugl-Meyer Assessment scale on lower limbs (FMA-L). The secondary indicators were the Berg balance scale (BBS), 10-meter walk test (10MWT), gait speed (GS), 6-min walk test (6MWT), functional ambulation category scale (FAC), timed up and go test (TUGT), Brunnstrom stage of lower limb function, weight-bearing, modified Barthel index (MBI), functional independence measure (FIM), stroke-specific quality of life questionnaire (SSQOL), World Health Organization quality of life assessment (WHOQOL), and National Institute of Health stroke scale (NIHSS).ResultsWe initially identified 343 relevant studies. Among them, 34 (totaling 2,008 patients) met the inclusion criteria. We found that patients treated with CIMT had significantly better primary indicator (FMA-L) scores than those not treated with CIMT. The mean differences were 3.46 (95% CI 2.74–4.17, P < 0.01, I2 = 40%) between CIMT-treated and conventional physiotherapy-treated patients, 3.83 (95% CI 2.89–4.77, P < 0.01, I2 = 54%) between patients treated with CIMT plus conventional physiotherapy and patients treated only with conventional physiotherapy, and 3.50 (95% CI 1.08–5.92, P < 0.01) between patients treated with CIMT plus western medicine therapy and those treated only with western medicine therapy. The secondary indicators followed the same trend. The subgroup analysis showed that lower extremity CIMT with device seemed to yield a higher mean difference in FMA-L scores than lower extremity CIMT without device (4.52, 95% CI = 3.65–5.38, P < 0.01 and 3.37, 95% CI = 2.95–3.79, P < 0.01, respectively).ConclusionCIMT effectively improves lower extremity motor dysfunction in post-stroke patients; however, the eligible studies were highly heterogeneous.Systematic review registration: https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=277466.