Longterm prognosis of analgesic withdrawal in patients with drug-induced headaches.

We studied long-term prognosis and prognostic variables for therapeutic outcome of analgesic withdrawal in 54 patients with drug-induced headaches. The duration of headache history was 21.9 +/- 12.8 years. Each patient took an average of 38.8 +/- 22.8 tablets or suppositories a week and an average of 2.5 distinct drugs. Most patients used drugs containing several components. Caffeine was contained in at least one drug in all cases, ergotamine in 80.0% and pyrazolone in 77.1%. All patients were admitted to the hospital for two weeks. The analgesics were discontinued abruptly and the withdrawal symptoms were alleviated by neuroleptics and neurotropics. During the second week of hospital stay we started a basic therapy with calcium antagonists or beta blockers in patients suffering from migraine initially and with tricyclic antidepressants, physical therapy or biofeedback in patients suffering from tension type headaches initially. At the end of the study (mean follow-up period = 16.8 +/- 13.6 months) 38 patients (70.1%) were evaluated. 76.3% of these patients had significantly reduced their analgesic intake, 60.5% had experienced a significant relief of headache both in intensity and frequency, and 23.7% were therapeutic failures. Analysis of the time course of relapse revealed the first six months after hospital discharge as the critical period determining long-term success. The variables tested for prognostic relevance (age, sex, duration of headache history, number of tablets or suppositories taken a week, organic mental syndrome, and type of initial headaches) were not statistically significant.