Objectives: To determine if the effectiveness of propranolol hydrochloride and amitriptyline hydrochloride are correlated with blood levels and/or with standardized test of pharmacologic effect and to determine which clinical variables are predictors of response to one or the other medication.
Design: Three-month modules of treatment with each drug and placebo in a randomized crossover design. Headache scores from daily diaries were calculated at monthly intervals, as were simultaneous blood levels of drug, supine and standing blood pressure, pulse rise with exercise, and salivary flow.
Setting: Outpatient headache clinic at the University of Kansas Medical Center, Kansas City.
Patients: Thirty consecutive patients with a history of frequent migraine.
Main outcome measurements: From headache scores, patients were classified as either propranolol responders, amitriptyline responders, or nonspecific responders. Clinical variables as predictors of response to medications were studied, as were effects on frequency, duration, and/or severity of headache.
Results and conclusions: No significant correlations were found between changes in headache score and blood level of drug or change in any of the physiologic measurements. Amitriptyline significantly reduced the severity, frequency, and duration of headache attacks; propranolol reduced the severity of attacks only. Amitriptyline response was correlated with female gender and baseline headaches of shortest duration and of highest frequency. Propranolol response was associated with attacks of greatest duration at baseline and with low pulse rise with exercise at baseline. Nonspecific response was associated with male gender and most frequent headaches by history.