Page 515 Acute Pain Management
P. 515
with neurological signs greatly increases the risk of an incomplete
recovery (Davies et al 2004)
Evidence from a single Electrical corruption of PCA pumps resulting in uncontrolled delivery of
case report or letter that syringe contents (Notcutt et al 1992)
directly affects patient The need for antisyphon valves when using PCA in order to prevent
safety cited as Level IV inadvertent delivery of an excessive dose of opioid (Kwan 1995)
Quality scoring
Where Cochrane reviews or reviews from other reputable sources were available, no
additional methodological quality evaluation was undertaken, and what was available in the
review was accepted as the quality scoring for these guidelines. For the remaining systematic
reviews containing controlled trials (Level I) and primary RCT evidence (Level II), methodological
quality was evaluated using a purpose‐built system, which identified studies that were
subsequently included in the review. The purpose‐built criteria included the presence of a
CONSORT diagram (Moher et al 2001), explicit inclusion and exclusion criteria, sound description
of the intervention and clear outcome measures. No quality evaluation was undertaken for
lower ranked evidence (Level III and Level IV), when this was the highest available level of
evidence. Thus this document is underpinned by the highest level, highest methodological
quality evidence available for each review question.
Conflicting evidence
If evidence was consistent, the most recent, highest level and highest quality references were
used. If it was conflicting, the same approach was taken (identifying highest level, highest
quality evidence) however examples were given of differences within the literature so that
readers could appreciate the ongoing debate. In some instances, particularly in acute pain
management in various patient populations, evidence was limited to case reports only, which
was made clear in the document as the best available evidence in this instance.
Management of retracted publications
In May 2009, two editorials (Shafer et al 2009; White et al 2009) were published in Anesthesia and
Analgesia giving details of 21 publications that had been retracted by a number of journals
because of allegations of scientific fraud. The editorial by Shafer (Shafer et al 2009) contains a
list of the retracted articles. This list can also be found at
http://www.aaeditor.org/HWP/Retraction.Notice.pdf.
The position of the journal was that unretracted articles ‘remain part of the unimpeached
literature, at least for now’. In a companion editorial White et al (White et al 2009) reviewed
both the retracted and unimpeached literature, ‘distinguishing our understandings that
remain fully supported from those that are no longer supported by the unimpeached
literature.’ Also in May 2009, Eisenach (Eisenach 2009) the editor of Anesthesiology, presented
a graph of numbers of citations of retracted and unretracted articles affected by this issue and
called for research re‐examining the conclusions of the retracted articles.
A July 2009 editorial by Neal (Neal 2009) described contact with ‘the lead or high ranking
authors’ of six original articles and one review article in that editor’s journal and which had not APPENDIX B
been retracted. These articles are listed in this editorial. He concluded that ‘Based on the
attestations of the involved coauthors and the investigations of the Chief Academic Officer of
Baystate Medical Center, the Editorial Board of Regional Anesthesia and Pain Medicine is
comfortable recommending that practitioners continue to make clinical decisions based on the
information contained within the seven below cited articles.’
Acute pain management: scientific evidence 467
