Aims: To evaluate, in an open trial, the pharmacotherapeutic efficacy of tricyclic antidepressant (TCA) drugs and gabapentin in patients with persistent myofascial pain and to identify patient and pain characteristics that may predict treatment outcome. Methods: A stepped pharmacotherapeutic protocol was employed. All 42 patients having persistent facial pain with tenderness of regional muscles were first prescribed amitriptyline, but those with side effects were subsequently transferred to nortriptyline. In patients where no response to TCAs was observed, gabapentin was initiated. Outcome was assessed by employing prospective diaries recording pain intensity measured with an 11-point (0-10) verbal pain scale (VPS). Individual characteristics in these patients and their influence on drug response and outcome were analyzed; specifically, patients treated with TCAs were compared with those subsequently treated with gabapentin. Chisquare and t tests were used to analyze the data. Results: A total of 23 patients responded to TCAs and continued on this regimen, while 19 were resistant to TCAs and were subsequently treated with gabapentin. Their mean (± SD) VPS score at baseline was 6.5 ± 1.9 on an 11-point scale. In TCA-treated patients, 43% showed ≥ 50% reduction in pain intensity. This was achieved with a mean amitriptyline dose of 16 ± 1.1 mg/d and a mean nortriptyline dose of 25 ± 2.1 mg/d. Patients who did not respond to TCAs were characterized by a significantly higher age, more comorbid medical illness, and evidence of more regional pain spread (P .05). In spite of not responding to TCAs, 36.8% of this group showed ≥ 50% reduction in pain intensity following gabapentin therapy at a mean daily dose of 973.7 ± 68.8 mg. Overall, a stepped approach employing TCAs and gabapentin resulted in 54.8% of all treated patients reporting improvements of ≥ 50% in VPS scores. Conclusion: This study has demonstrated the good pharmacotherapeutic response of persistent myofascial pain, even in more severe cases. Not being a randomized controlled trial, the results may be biased and should be interpreted with caution. Patients who do not respond to TCAs may be a distinct subgroup and this needs further investigation. The results also suggest that gabapentin, at a lower dose than previously reported, is a good alternative in TCA-resistant patients. Keywords: amitriptyline, antiepileptic drug, myalgia, nortriptyline, orofacial pain