Hepatocyte growth factor (HGF) is a potent angiogenic factor. The efficacy and safety of intramuscular injection of a naked plasmid encoding human HGF gene (beperminogene perplasmid, Collategene) was investigated in patients with critical limb ischemia (CLI) in a multicenter, randomized, double-blind, placebo-controlled trial. The randomization ratio for plasmid to placebo was 2:1. Injection sites were selected in each patient limb based on angiographic findings. Placebo or plasmid was injected on days 0 and 28. Evaluation of efficacy was carried out after 12 weeks. The primary end point was the improvement of rest pain in patients without ulcers (Rutherford 4) or the reduction of ulcer size in patients with ulcer(s) (Rutherford 5). Secondary end points were ankle-brachial pressure index, amputation, and quality of life (QOL). Forty-four patients were treated, and we performed interim analysis of efficacy in 40 patients. The overall improvement rate of the primary end point was 70.4% (19/27) in HGF group and 30.8% (4/13) in placebo group, showing a significant difference (P=0.014). In Rutherford 5 patients, HGF achieved a significantly higher improvement rate (100% [11/11]) than placebo (40% [2/5]; P=0.018). HGF plasmid also improved QOL. There were no major safety problems. HGF gene therapy is safe and effective for CLI.