Skip navigation

Treatment of Medication-Induced Headache

Clinical bottom line

Information concerning the treatment of medication-induced headache (MIH) is scant and of poor methodological quality. Evidence based on mainly uncontrolled studies indicates that complete withdrawal of overused medications is required for positive long-term benefit. There is a lack of high quality evidence evaluating the role of any therapeutic treatment in the management of MIH.

Systematic review:

Zed PJ, Loewen PS, Robinson G. Medication-induced headache: Overview and systematic review of therapeutic approaches. Neurology. 1999; 33: 61-72

  • Date review completed: June 1998
  • Number of trials included: three randomised controlled trials
  • Number of patients: 228 patients in active and control groups
  • Control groups: placebo, amitriptyline, symptomatic medications
  • Main outcomes: Headache Index (HI: product of duration and severity), headache frequency
Inclusion criteria were any reports of management of medication-induced headache (MIH) with any study design Reviewers conducted a search for English language publications on MEDLINE and reference lists of retrieved reports. All identified studies were qualitatively evaluated regardless of study design. Reviewers provided a descriptive summary of included studies.


Nineteen studies were included, the majority of which were uncontrolled studies or case series. Three were randomised controlled trials (RCT). Although of higher quality than the uncontrolled studies, these three trials also had methodological shortcomings. Two of the trials were not blind and the other had a total of only six patients. The results for these three RCTs are summarised below.

One double blind RCT of six migraine patients who were ergotamine overusers compared rapid withdrawal of headache drugs and 4 mg subcutaneous (SC) sumatriptan versus placebo. All sumatriptan treated patients had complete headache relief for six to 12 hours, versus none in placebo group.

One unblinded RCT compared four interventions in 200 patients with MIH due to drugs including paracetamol, aspirin, barbiturates, opiates, sedatives and muscle relaxants. The treatment groups were:

1. abrupt discontinuation of analgesics

2. abrupt discontinuation of analgesics plus amitriptyline 50 mg per day

3. continued analgesic use

4. continued analgesic use plus amitriptyline 50 mg per day.

At four weeks the most effective intervention was abrupt withdrawal plus amitriptyline with 72% reduction in HI. Abrupt withdrawal was more effective (60% reduction in HI) than continued use of usual meds, and amitriptyline (45% reduction in HI) which was better than continued use of usual meds alone (25% reduction in HI). There was a high dropout rate (62%) in both abrupt withdrawal groups therefore the reported benefit of this treatment is based on a small minority of the patients.

One unblinded RCT compared naproxen 500 mg twice daily with symptomatic medications only (antiemetics, simple analgesics, hydration) in 22 patients with MIH due to ergotamine overuse. The naproxen group had less headache over eight days (p<0.001), however at follow-up there was no difference between groups for HI. Three patients on naproxen withdrew due to adverse effects.


This does not amount to a significant body of evidence. Few trials, relatively poor quality, and few patients.