Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: NSAIDs and muscle relaxants reduce acute low-back pain; manipulation, back schools, and exercise reduce chronic low-back pain

ACP J Club. 1998 May-June;128:65. doi:10.7326/ACPJC-1998-128-3-065

Source Citation

van Tulder MW, Koes BW, Bouter LM. Conservative treatment of acute and chronic nonspecific low back pain: a systematic review of randomized controlled trials of the most common interventions. Spine. 1997 Sep 15;22:2128-56.



To evaluate the effectiveness of the most common conservative treatment options for acute and chronic low-back pain.

Data sources

Studies were identified using MEDLINE (1966 to September 1995), EMBASE (1980 to September 1995), and PsycLIT (1984 to September 1995) using the terms low back pain, backache, musculoskeletal diseases, joint diseases, spinal diseases, and physical therapy and the names of specific interventions. Bibliographies of retrieved articles were scanned for additional articles.

Study selection

English-language, randomized controlled trials were selected if they evaluated the effectiveness of analgesics, nonsteroidal anti-inflammatory drugs (NSAIDs), muscle relaxants, antidepressants, epidural steroid injections, transcutaneous electrical nerve stimulation, traction, behavior therapy, orthoses, electromyographic biofeedback, or acupuncture in nonsurgical patients with acute (persisting ≥ 12 wk) low-back pain. Studies of cervical pain were excluded.

Data extraction

Trials were assessed for methodologic quality (17 criteria). Data were extracted on type of therapy, number of patients, pain intensity, overall improvement, functional status, and whether the study conclusion was positive (the experimental therapy was more effective) or negative (no difference was shown between the therapies studied, or the reference therapy was more effective).

Main results

150 trials (68 on acute low-back pain, 81 on chronic low-back pain, and 1 on both) met the selection criteria. 35% of acute and 25% of chronic low-back pain trials were of high methodologic quality. Results were reported on the basis of the level of evidence; strong evidence referred to multiple high-quality trials.

For acute low-back pain, strong evidence showed that NSAIDs were more effective than placebo but not more effective than analgesics. Different NSAIDs were equally effective. Similarly, strong evidence showed that muscle relaxants were more effective than placebo and that all types were equally effective. In contrast, strong evidence showed that bed rest and exercise were not effective for acute low-back pain.

For chronic low-back pain, strong evidence showed that manipulation, intensive back school programs in an occupational setting, and exercise were all effective.


Few trials of therapies for low-back pain have high methodologic quality. NSAIDs and muscle relaxants are effective for acute low-back pain. Manipulation, back schools, and exercise are effective for chronic low-back pain.

Source of funding: Dutch Health Insurance Board.

For article reprint: Dr. M.W. van Tulder, Institute for Research in Extramural Medicine, Faculty of Medicine, Vrije Universiteit, Van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands. FAX 31-20-44-48262.


Almost everyone has low-back pain at some point in his or her life. Although the pain is usually transient and self-limiting, its management presents major medical challenges. van Tulder and colleagues used a grading system for quality that was developed for the U.S. Agency for Health Care Policy and Research (1) and concluded that most back pain studies are inadequate.

For acute back pain, although several different therapies are useful, no intervention may be needed and simple self-medication can often suffice. Chronic back pain is different, and investigation to determine the underlying cause may lead to appropriate therapy. The optimal approach to managing back pain remains elusive. Therefore, defining the appropriate outcome measures in studies of several interventions has become increasingly important. The authors recommend back schools and other nondrug treatments for chronic pain. Bed rest is ineffective. No major study seems to have addressed yoga, which is becoming popular, or chiropractic “adjustment,” which is considered effective but not necessarily cost-effective for acute pain. As usual, the large number of approaches testifies to the ultimate inadequacy of any one treatment or perhaps to our failure to understand the complexity of back pain.

Nonspecific low-back pain is so common that physicians are well advised not to overtreat it. Physicians also recognize that trials of therapy can be misleading in the management of the individual patient, recalling Dubos's dictum that the measurable drives out the important. The customary risk-benefit equation measures the benefits and risks of the intervention but fails to address the risk for failing to treat. For nonspecific low-back pain, associated risk seems to be minimal; therefore, to be acceptable, the risks introduced by treatment should be small. This is important because no single treatment has emerged that can be considered more than palliative.

George E. Ehrlich, MD
Low Back Pain InitiativeWorld Health OrganizationPhiladelphia, Pennsylvania, USA

George E. Ehrlich, MD
Low Back Pain InitiativeWorld Health Organization
Philadelphia, Pennsylvania, USA


1. Bigos S, Bowyer O, Breen G, et al. Acute low back problems in adults. Clinical Practice Guideline No. 14; AHCPR Publication No. 95-0642. Rockville, MD: Agency for Health Care Policy and Research, Public Health Services, U.S. Department of Health and Human Services; Dec. 1994.