This prospective, double-blind, randomized, sham-controlled trial was designed to control for patient and investigator bias in assessing symptomatic improvement after percutaneous myocardial laser revascularization (PMLR) therapy. Eighty-two patients with stable angina pectoris (class III or IV) not amenable to conventional revascularization and with evidence of reversible ischemia, ejection fraction >/=25%, and myocardial wall thickness >/=8 mm were randomized to either PMLR with optimal medical therapy (n = 40) or to a sham procedure with optimal medical therapy (n = 42). With the exception of 1 laser technician, all patients, investigators, and assessors were blinded to treatment through the 12-month follow-up. The primary end point was restricted to Canadian Cardiovascular Society angina class improvement to limit the number of patients exposed to a sham procedure. Secondary assessments included medication usage, quality of life, exercise testing, ejection fraction, and hospitalizations. The incidence of serious adverse events, as determined by cardiac event-free survival at 12 months, was similar between groups. At 12 months, Canadian Cardiovascular Society angina scores improved by >/=2 classes in significantly more PMLR-treated patients than sham control patients (35% vs 14%, p = 0.04). Angina-specific quality-of-life measures were significantly higher in the PMLR group at each follow-up (p <0.05). Exercise and medication usage was similar between groups at 12 months. We conclude that PMLR therapy is reasonably safe and effective as symptomatic improvement in patients refractory to medical therapy, and that the clinical benefit is not attributable to placebo effect or investigator bias.