Health care professionals depend on good sources of information to keep up-to-date on evidence-based medicine. One of the most reliable publications comes from the Cochrane Collaboration. Everyone pays attention when they publish a study.
The Cochrane Collaboration is a group of over 11,500 volunteers who systematically review all relevant randomized controlled trials on a specific topic. The people in the group are located in more than 90 countries. They also study the results of non-randomized, observational studies. The results of these systematic reviews are published in the Cochrane Library.
In this study, the use of nonsteroidal antiinflammatory drugs (NSAIDs) for low back pain (LBP) was reviewed. The goals were to 1) look for effects of NSAIDs and 2) find out which type of NSAID works the best. Specifically, COX-2 inhibiting NSAIDs were considered. These drugs have a lower risk of gastrointestinal side effects. For this reason, they have recently become more popular than the traditional NSAIDs.
The Cochrane reviewers searched for studies done on low back pain with NSAIDs. They found 28 trials that were of acceptable quality to be included. NSAIDs were compared with placebos, Tylenol, muscle relaxants, and other drugs (such as narcotics). Some studies compared traditional NSAIDs with the newer COX-2 inhibitors.
Patients included were adults age 18 and older with nonspecific low back pain with or without sciatica. Results were measured using pain, function, return to work status, and number of days off work. Other factors reviewed included side effects of the drugs, visits to a health care specialist, and general function. Range of motion, strength, and degrees of straight leg raising were also noted.
The reviewers used a rating system of levels of evidence to summarize the results of the studies. Outcomes were reported on a scale from no evidence to strong evidence. Moderate, limited, and conflicting evidence were in between on the continuum.
Any concerns about the way the studies were conducted or the design of the study were reported by the group. For example, follow-up was considered inadequate in more than half of the trials. Random selection and treatment groups were not always concealed as they should be to qualify as random-controlled trials. And in some studies, it simply wasn’t clear if the medication was being taken as prescribed.
When the data is of good quality, then results can be pooled together. This gives a much larger subset of results from which to make conclusions. In this meta-analysis, 11 studies on NSAIDs with placebo groups for low back pain qualified for pooling data. The results showed:
Some of the studies were just too small to come to any firm conclusions. For example whether or not NSAIDs work better when given as an intramuscular injection, capsule, or gel remains largely unknown. And there may be a bigger role for the use of B vitamins in acute and/or chronic back pain, but there hasn’t been enough evidence to come to any clear conclusions.
The authors suggest better studies are needed to really evaluate the use of NSAIDs. This is true both for patients with acute back pain as well as those suffering from chronic low back pain. It would also be helpful to compare dose, effectiveness, and side effects for various subgroups of patients.