Context Various guidelines recommend different strategies for selecting and
sequencing acute treatments for migraine. In step care, treatment is escalated
after first-line medications fail. In stratified care, initial treatment is
based on measurement of the severity of illness or other factors. These strategies
for migraine have not been rigorously evaluated.
Objective To compare the clinical benefits of 3 strategies: stratified care, step
care within attacks, and step care across attacks, among patients with migraine.
Design and Setting Randomized, controlled, parallel-group clinical trial conducted by the
Disability in Strategies Study group from December 1997 to March 1999 in 88
clinical centers in 13 countries.
Patients A total of 835 adult migraine patients with a Migraine Disability Assessment
Scale (MIDAS) grade of II, III, or IV were analyzed as the efficacy population;
the safety analysis included 930 patients.
Interventions Patients were randomly assigned to receive (1) stratified care (n =
279), in which patients with MIDAS grade II treated up to 6 attacks with aspirin,
800 to 1000 mg, plus metoclopramide, 10 mg, and patients with MIDAS grade
III and IV treated up to 6 attacks with zolmitriptan, 2.5 mg; (2) step care
across attacks (n = 271), in which initial treatment was with aspirin, 800
to 1000 mg, plus metoclopramide, 10 mg. Patients not responding in at least
2 of the first 3 attacks switched to zolmitriptan, 2.5 mg, to treat the remaining
3 attacks; and (3) step care within attacks (n = 285), in which initial treatment
for all attacks was with aspirin, 800 to 1000 mg, plus metoclopramide, 20
mg. Patients not responding to treatment after 2 hours in each attack escalated
treatment to zolmitriptan, 2.5 mg.
Main Outcome Measures Headache response, achieved if pain intensity was reduced from severe
or moderate at baseline to mild or no pain at 2 hours; and disability time
per treated attack at 4 hours for all 6 attacks, compared among the 3 groups.
Results Headache response at 2 hours was significantly greater across 6 attacks
in the stratified care treatment group (52.7%) than in either the step care
across attacks group (40.6%; P<.001) or the step
care within attacks group (36.4%; P<.001). Disability
time (6 attacks) was significantly lower in the stratified care group (mean
area under the curve [AUC], 185.0 mm · h) than in the step care across
attacks group (mean AUC, 209.4 mm · h; P<.001)
or the step care within attacks group (mean AUC, 199.7 mm · h; P<.001). The incidence of adverse events was higher
in the stratified care group (321 events) vs both step care groups (159 events
in across-attack group; 217 in within-attack group), although most events
were of mild-to-moderate intensity.
Conclusion Our results indicate that as a treatment strategy, stratified care provides
significantly better clinical outcomes than step care strategies within or
across attacks as measured by headache response and disability time.