The latest evidence from major systematic reviews
Dr Chris Cates discusses new Cochrane reviews relevant to
he Cochrane Library consists of a regularly updated collection of evidence-based medicine databases which are available online or on CD.
The latest issue (2002, Issue 2) contains several new reviews that are relevant to primary care. The first relates to selective b-blockers and chronic obstructive pulmonary disease (COPD).
?-blockers and COPD
In general, b-blockers have been shown to be useful in secondary prevention for patients with ischaemic heart disease, but problems arise because many of these patients have smoked and also have COPD, so we may be reluctant to consider a b-blocker in case this makes the chest problems worse.
The new review suggests cardioselective b-blockers do not cause important deterioration in COPD and should not be withheld in patients with COPD1.
The main results were as follows. Eleven studies of single-dose treatment and eight of treatment for longer duration, ranging from two days to 12 weeks, met selection criteria. Cardioselective b-blockers, given as a single dose or for longer duration, produced no significant change in FEV1 or respiratory symptoms compared with placebo, and did not significantly affect the FEV1 treatment response to
b2-agonists. A subgroup analysis revealed no significant change in results for those participants with severe chronic airways obstruction or for those with a reversible obstructive component.
The reviewers concluded that cardioselective b-blockers given to patients with COPD do not produce adverse respiratory effects. Given their demonstrated benefit in conditions such as heart failure, coronary artery disease and hypertension, cardioselective
?-blockers should not be withheld from patients with COPD.
Treatments for tennis elbow
A review on tennis elbow2 has shown topical NSAID treatment to be effective in the short-term relief of pain (although this comes from only 150 patients randomised) and even smaller numbers of patients from two trials suggested that patients were more likely to show subjective improvement with symptoms (in the short-term) following a steroid injection than oral NSAID.
More work is needed to see how topical NSAID treatment compares with oral NSAID and to study the outcomes for longer than four weeks. A note of caution on the comparison between steroid injection and oral NSAID: as the outcome was subjective I would be concerned about bias in trials with no dummy injection.
There was only one study (Saartok 1986) that did use a dummy injection and this did not show a significant difference from NSAID, whereas the other study (Hay 1999) with no dummy injection did show a significant benefit.
A larger double-blind, double-dummy study with a head-to-head comparison is therefore needed to confirm whether there is a real difference between these two approaches (see forest plot below).
Other new reviews
Other new reviews have found no important differences between carbamazepine and phenytoin as monotherapy for epilepsy, no good evidence to recommend dehumidifiers in asthma, and interestingly only one randomised controlled trial on 22 patients was found comparing surgical with non-surgical treatment of carpal tunnel syndrome. More research needed!
Comparison: NSAID versus steroid injection
Outcome: no improvement on patient's self assessment at four weeks
Study Experiment (n/N) Control (n/N) Odds ratio (95% CI) Weight (%) Odds ratio (95% CI)
Hay 1999 23/53 4/52 79.0 6.37 [2.66, 15.22]
Saartok 1986 4/10 5/11 21.0 0.81 [0.15, 4.38]
Total (95% CI) 27/63 9/63 100.0 4.13 [1.90, 8.95]
0.1 0.2 1 5 10
Favours NSAID Favours injection
1. Salpeter S et al. Cardioselective ?-blockers for chronic obstructive pulmonary disease (Cochrane Review). In: Cochrane Library (Second edition) Oxford: Update Software, 2002
2. Green S et al. Non-steroidal anti-inflammatory drugs (NSAIDs) for treating lateral elbow pain in adults (Cochrane Review). In: Cochrane Library (Second edition): Oxford: Update Software, 2002