Neurally mediated syncope is a complex syndrome that is often difficult to manage using currently available treatment strategies. The efficacy and safety of clonazepam was evaluated in 35 patients with refractory neurally mediated syncope. All patients had syncope (n = 33) or disabling presyncope (n = 2) and a positive head-up tilt table test (HUTT) despite treatment with one or more of the following therapies: beta-blocker, high-salt diet, fludrocortisone, elastic compression stockings, and disopyramide. Clonazepam was initiated at 0.5 mg/day and titrated in 0.25-0.5 mg/day increments for symptom control. Early (first 8 weeks) symptomatic response was achieved in 31 of 35 (89%) patients. Early HUTT reverted to negative in 29 of 35 (83%) patients. Two patients discontinued clonazepam during early follow-up due to side effects. Thirty-three patients received long-term clonazepam therapy. Twenty-five patients had late HUTT with 21 remaining negative. Of the eight patients who did not have late HUTT, one patient discontinued clonazepam prior to HUTT due to side effects. Seven patients refused late HUTT. All seven patients achieved symptomatic control on clonazepam with two requiring dose titration. Of the 21 patients with a negative late HUTT, 18 achieved symptomatic control with two of these patients requiring dose titration. Two patients who had only partial symptom control despite dose titration achieved total symptomatic control with the addition of disopyramide and beta-blockers. Two patients with a negative late HUTT discontinued clonazepam due to side effects. One patient had been symptomatically controlled while the other had recurrent symptoms with dose limiting side effects occurring after clonazepam dose titration. In the 4 patients with a positive late HUTT, 2 patients were symptomatically controlled, 1 patients required combination therapy with a beta-blocker to achieve symptomatic control, and 1 patient discontinued therapy due to side effects. Overall, 29 of 35 (83%) patients continue to receive clonazepam with symptom control. Based on intention-to-treat HUTT results, 21 of 35 (60%) patients were responders. Four patients required clonazepam dose titration and three required combination therapy with clonazepam plus disopyramide and/or a beta-blocker to achieve control. Clonazepam was discontinued in 6 patients, 5 for side effects and 1 following a transient ischemic attack. Clonazepam appears to be an effective therapeutic alternative in patients with refractory neurally mediated syncope. Based on our preliminary findings, a placebo controlled evaluation of clonazepam in neurally mediated syncope is warranted.