Question: In patients with acute low-back pain, is 4 days of bed rest more effective than continued normal daily activity, taking into account a patient's type of work?
Design: Randomized (allocation concealed), unblinded, controlled trial with a 3-month follow-up.
Setting: Practices of 54 rheumatologists and general practitioners.
Patients: 278 ambulatory patients who were 18 to 65 years of age (mean age, 44 years; 53% women) with acute low-back pain or a recent episode (<72 hours) of chronic low-back pain with spontaneous lumbar pain rated =40 mm on a 100-mm visual analog scale (VAS). Exclusion criteria were pain radiating distal to the buttocks; compressive, posttraumatic, inflammatory, infectious, or tumoral disease; or low-back pain resulting from an occupational accident. 98% of the patients had complete follow-up.
Intervention: Patients were stratified by their type of work (none, physical, or sedentary) and were allocated to treatment with 4 days of bed rest (
n = 138) or to continuation of normal daily activities as far as the pain allowed (
n = 140). Patients in the bed-rest group were to stay in bed except for personal care and meals and needed to spend =16 of every 24 hours in bed. All activity was forbidden in the first 4 days. Patients in the activity group could not spend >12 of every 24 hours in bed. All patients received acetaminophen (1 g, 3 times/day) and thiocolchicoside (8 mg, 2 times/day). Nonsteroidal anti-inflammatory drugs were permitted in cases of real need.
Main outcome measures: The main outcome was back pain on day 6 or 7, assessed with the 100-mm VAS. Secondary outcomes were back pain at 1 and 3 months, functional disability assessed with the Eifel index (French translation of the Roland-Morris index), stiffness (the Schober test), and the number of sick-leave days taken.
Main results: Analysis was by intention to treat. 72% of the patients in the bed-rest group and 90% in the normal-activity group complied with the prescribed treatment. At day 6 or 7, the groups did not differ with respect to pain intensity, functional disability, or vertebral stiffness (
table ) and also did not differ with respect to these outcomes at 1 and 3 months. Initial sick leave was taken by more patients in the bed-rest group than by those in the normal activity group (86% vs 52%, p < 0.001). The difference between the bed-rest and normal-activity groups with respect to prescription rates for sick leave was greater for patients with a sedentary occupational activity (76% vs 25%) than for those with a physical occupational activity (94% vs 69%).
Conclusions: In patients with acute low-back pain, 4 days of bed rest was no more effective than continued normal daily activity at 1 week or at 1 or 3 months. The prescription for bed rest increased the prescription rate for sick leave in patients with sedentary occupational activities.