DARE: Search Results
Document: 26
Conservative treatment of acute and chronic nonspecific low back pain: a
systematic review of randomized controlled trials of the most common
interventions.
van Tulder M W, Koes B W, Bouter L M..
Conservative treatment of acute and chronic nonspecific low back pain: a
systematic review of randomized controlled trials of the most common
interventions.
Spine,
1997,
22(18),
pp.2128-2156
Record status
This record is a structured abstract written by CRD reviewers. The original
has met a set of quality criteria. Since September 1996 abstracts have been
sent to authors for comment. Additional factual information is incorporated
into the record. Noted as [A:....].
Author's objective
To assess the effectiveness of the most common conservative types of treatment
for patients with acute and chronic nonspecific low back pain (LBP). Type
of intervention
Treatment.
Specific interventions included in the review
Bed rest, orthoses, exercise therapy, back schools, spinal manipulation,
analgesics, non-steroidal anti-inflammatory drugs (NSAIDS), muscle relaxants,
antidepressants, epidural steroid injections, transcutaneous electrical nerve
stimulation (TENS), traction, behavioural therapy, electromyographic
biofeedback, and acupuncture. Control treatments varied between studies, and
included placebo for some evaluations.
Participants included in the review
Patients with acute (low back pain of 6 weeks or less duration), subacute
(duration 6-12 weeks), or chronic low back pain (duration 12 weeks or more).
People with cervical back pain or a combination of thoracolumbar and cervical
back pain were excluded unless the results for thoracolumbar back pain were
presented separately. Surgical patients were also excluded.
Outcomes assessed in the review
Pain, functional status, overall improvement.
Study designs of evaluations included in the review
Randomised controlled trials (RCTs), published as full reports. Unpublished
studies, and those with quasi-randomisation procedures, were excluded.
What sources were searched to identify primary studies?
MEDLINE (1966-September 1995), EMBASE drugs and pharmacology database
(1980-September 1995), and PsycLIT (1984- September 1995) (search terms
provided). The bibliographies of identified studies were examined. Only
English language studies were selected.
Criteria on which the validity (or quality) of studies was assessed
Each study was assessed on study population (homogeneity, baseline
comparability, randomisation, drop-outs), interventions (standardisation,
description, controls, avoidance of co-interventions, placebo controlled),
effect (blinding of patients and assessors, relevance of outcome measures,
follow-up), data presentation and analysis (intention-to-treat analysis,
frequencies of important outcomes), and compliance (drug trials only). Each
criterion was weighted, and each study could score a potential maximum of 100
points, with higher scores indicating higher methodological quality. Studies
were also attributed one of four ratings for level of evidence (strong,
moderate, limited, none), dependent upon quality and outcome, according to US
Clinical Practice Guideline for Acute Low Back Problems in Adults.
How were the judgements of validity (or quality) made?
Assessment was performed by 2 independent reviewers (not blinded to source or
outcome of trials) and disagreements were resolved through discussion or
recourse to a third reviewer.
How were decisions on the relevence of primary studies made?
Titles and abstracts of studies were examined. The process of decision making
was not described (i.e. whether independent assessment of material was carried
out by more than one reviewer, and methods used for resolving disagreements).
How was the data extracted from primary studies?
Data were extracted relating to the outcomes of pain, functional status, and
overall improvement. A study was considered to have a positive result if the
experimental intervention was more effective compared with control for at
least one of these outcomes. Studies were considered to have negative results
if there were no differences between groups on any of the outcomes, or if the
control treatment proved to be more effective for at least one of the
outcomes. If the experimental intervention was more effective for one outcome
measure, but less effective for another, or if these outcome measures were not
assessed, the study was categorised as "no conclusion". No information was
provided on the process by which data were extracted (i.e. whether done
independently by more than one reviewer).
Number of studies included
Overall, 150 RCTs were included. Of these, 68 recruited people with acute
LBP, 81 chronic LBP, and one RCT included participants with both acute and
chronic LBP. Numbers recruited to individual studies are not given in all
cases, but it appears that at least 10,000 participants were included in this
review.
How were the studies combined?
By narrative summary and tabulation.
How were differences between studies investigated?
Distinction was made between studies of high methodological quality (50 points
or more) and low quality (less than 50 points).
Results of the review
Overall, the methodological quality of included studies was low, but the
methodological quality of drug trials was generally higher compared with
trials of other conservative treatments.
Interventions for acute low back pain (LBP):
There is moderate evidence that analgesics are equivalent to non-steroidal
anti-inflammatory drugs (NSAIDS), but no evidence that analgesics are more
effective than electroacupuncture or ultrasound. There is strong evidence to
show that NSAIDS are more effective than placebo for non-sciatic LBP, that
NSAIDS are equivalent to analgesics, and that various types of NSAIDS are
equivalent. There is strong evidence that muscle relaxants are more effective
than placebo, and that different types of muscle relaxants are equivalent.
Bed rest is not effective. Exercise therapy is no more effective than other
conservative treatments. There were contradictory results for the
effectiveness of back school. There was limited evidence that manipulation
and for traction were more effective than placebo. There was limited evidence
that epidural steroid injections were more effective compared with
subcutaneous injections of lignocaine. There was no evidence to show that
transcutaneous electrical nerve stimulation (TENS) or behaviour therapy were
more effective than other conservative treatments.
Interventions for Chronic LBP:
In terms of analgesics, paracetamol is equivalent to diflunisal. Muscle
relaxants show a positive result when compared with placebo. Antidepressants
proved to be no more effective than placebo. There is moderate evidence for
the effectiveness of NSAIDS. Epidural steroid injections proved to be more
effective than placebo, but results of comparisons with injections of local
anaesthetic or muscle relaxant were contradictory. Manipulation is more
effective than placebo, usual general practitioner (GP) care, bed rest,
analgesics, and massage. Back school in an occupational setting may be more
effective than no treatment, but it is unclear whether back school is more
effective than other conservative treatments. Electromyographic biofeedback
is no more effective than waiting list, placebo, or other conservative
treatments. Exercise therapy appears to be more effective than other
conservative treatments, but no specific type of exercise proved to be more
effective than another. No significant differences were observed between 2
different types of traction. For orthoses, there were no significant
differences for corsets with or without lumbar support. Behavioural therapy
may be more effective compared with waiting list or other types of
conservative treatment. However, no one type of behavioural therapy emerges
as being superior to others. TENS does not appear to be any more effective
compared with waiting list, placebo, or other conservative treatments. There
is no evidence to show that acupuncture is more effective compared with
placebo, waiting list controls, or other conservative treatments.
Was any cost information reported?
No.
Author's conclusions
Many therapeutic interventions are available for, and used in the treatment
of, acute and chronic LBP. The quality of the design, execution, and
reporting of RCTs should be improved to establish strong evidence for the
effectiveness of the various therapeutic interventions for acute and chronic
LBP.
CRD commentary
Study selection, quality assessment, presentation of primary material, and
combining of data, are all well conducted in this review. The search strategy
includes only published English language studies, and there is no mention of
accessing databases specialising in complementary medicine, which may mean
that relevant material has been missed. There is limited information on the
processes of data extraction and decisions for including/excluding studies.
This review is an update of an earlier review.
What are the implications of the review?
See author's conclusion.
Subject index terms
Subject indexing assigned by NLM: Low-Back-Pain/th [therapy] Acute-Disease;
Chronic-Disease; Databases,-Bibliographic; Randomized-Controlled-Trials;
Treatment-Outcome
Correspondence address
Dr. M M van Tulder, Institute for Research in Extramural Medicine, Vrije
Universiteit, Amsterdam, The Netherlands.
URL
Copyright
University of York, 1999
Database no.: DARE-971202
