Reason for withdrawal from publication
At August 2015, this review has been withdrawn. It is correct at the date of publication, and previous versions can be accessed in the ‘Other versions’ tab on the Cochrane Library. The Cochrane Editorial Unit (CEU) agreed with the authors of the feedback that the review was misleading, and because the original author team was unavailable to update the review, the CEU advised that it should be withdrawn. See below for full details. PaPaS is seeking a new author team to develop a new review which will serve to update the original.
Feedback 1, received 15 February 2015
Dr Vicente Ruiz Garcia vicenteruizgarcia@gmail.com
With colleagues Xavier Bonfill Cosp, Eduardo Lopez Briz, Rafa Carbonell, Jose Luis Gonzalvez Perales, Sylvia Bort Martí, and Marta Roque Figuls.
Comment: Dear editor:
We have read the update of Zeppetella and Davies about management of breakthrough pain in cancer patients (1). We think that this review is very important to help clinicians and patients to decide whether the new treatments for the breakthrough pain in cancer could be a useful alternative to morphine.
In this update, some comments that we made to the previous review (the letter was not published) (2) were considered by the authors; in particular, not pooling the results obtained for oral transmucosal fentanyl citrate (OTFC) versus placebo, with those of OTFC versus morphine, and those of two trials that were titration of doses of fentanyl. However, we do not agree with the authors’ results when they state: “When compared with placebo or oral morphine, participants gave lower pain intensity and higher pain relief scores for transmucosal fentanyl formulations at all time points”. First, the outcomes at 15 min (the most important to obtain a quick relief of pain), Pain intensity Difference (Comparison 2. Transmucosal opioid versus oral morphine) and Comparison 4. OTFC versus intravenous morphine), failed to show statistically significant differences with oral morphine (mean difference 0.37 CI 95% 0.00‐0.73) and with morphine iv (mean difference 0.80 CI 95% 0.00‐1,60). In any case the results had no clinical relevance. Moreover the authors state “at all time points”, whereas they do not provide any data for longer times (i.e. 30, 45, 60 min). In addition, authors state “transmucosal fentanyl citrate are safe (..) (compared with both placebo and morphine) in relieving breakthrough pain”. Surprisingly no analysis of adverse events have been done that were only described in each study. Seven out 15 were crossover trials and it was impossible to draw conclusions about it.
As reviewers we know that multiple comparisons could be made, but the most clinical interesting comparison is the gold standard, morphine.
The review only shows in SOF, comparisons of fentanyl with placebo and concludes, that it is effective. Surprisingly, there is no mention of morphine comparison, which we consider a key point, because no patient will take placebo if he has a breakthrough pain, but morphine for obvious reasons.
In our opinion, traversing the authors’ conclusions, this review did not show that the use of oral and nasal transmucosal fentanyl is an effective alternative to morphine for patients with breakthrough cancer pain.
Reply
The authors of the review were contacted but chose not to provide a response to the feedback.
Contributors
Kate Seers, PaPaS Feedback Editor, and Anna Hobson, PaPaS Managing Editor.
Additional feedback 2, received May 2015
On 15 February 2015, Dr Vicente Ruiz Garcia (University Hospital La Fe, Spain), and his colleagues Xavier Bonfill Cosp, Eduardo Lopez Briz, Rafa Carbonell, Jose Luis Gonzalvez Perales, Sylvia Bort Martí, and Marta Roque Figuls submitted feedback via the Cochrane Library. The main complaint is available above.
On 23 February, 16 March and 15 April, Kate Seers (Feedback Editor, PaPaS) contacted the authors and invited them to respond. On 21 March, John Zeppetella (lead author) declined to provide a formal response. No response was received from Andrew Davies (second author).
On 31 March, PaPaS sought advice from the Cochrane Editorial Unit (CEU) on how to manage the issue; advised to publish without a response, depending on nature of feedback.
On 11 May 2015, review re‐published with feedback incorporated.
On 20 May 2015, Marta Roqué Figuls (Statistician, Iberoamerican Cochrane Centre) wrote to the CEU repeating the initial claim. They did not agree with the approach decided upon by the Co‐Ed, which was supported by the EiC. They stated “The CCIb assessment is that the review presents methodological shortcomings, and the conclusions are skewed in favor of fentanyl. Consequently, we support Vicente and colleagues’ petition to re‐assess the publication status of the review.”
Reply
On 9 June, Christopher Eccleston (Co‐ordinating Editor, PaPaS) advised that the review remain unchanged until it was ready for updating in September 2015.
On 25 June 2015 David Tovey (Editor in Chief, Cochrane) responded to say “We have now had a chance to appraise this review and also obtain a report from our screening team… In summary we agree with almost all of the criticisms made of the review, and are concerned that the flaws may mean that the findings are misleading as currently presented. We note that the authors have declined to respond to the useful comments provided by Vicente. Having considered this and discussed it internally, we agree with our colleagues at the IbCC that the review should be withdrawn temporarily until the errors have been fixed and the review updated. We would like to re‐screen the review before any update or amendment is published.”
On 2 July and 23 July, Anna Hobson (Managing Editor, PaPaS) again invited the authors to respond to the initial feedback and subsequent reviews by 23 July. No response was forthcoming.
At August 2015, the review was withdrawn.
Contributors
Kate Seers (Feedback Editor, PaPaS), Cochrane Editorial Unit (CEU), Christopher Eccleston (Co‐ordinating Editor, PaPaS), David Tovey (Editor in Chief, Cochrane), Anna Hobson (Managing Editor, PaPaS).
The editorial group responsible for this previously published document have withdrawn it from publication.
Feedback
Feedback received, 15 February 2015
Summary
Dr Vicente Ruiz Garcia vicenteruizgarcia@gmail.com
Comment: Dear editor:
We have read the update of Zeppetella and Davies about management of breakthrough pain in cancer patients (1). We think that this review is very important to help clinicians and patients to decide whether the new treatments for the breakthrough pain in cancer could be a useful alternative to morphine.
In this update, some comments that we made to the previous review (the letter was not published) (2) were considered by the authors; in particular, not pooling the results obtained for oral transmucosal fentanyl citrate (OTFC) versus placebo, with those of OTFC versus morphine, and those of two trials that were titration of doses of fentanyl. However, we do not agree with the authors’ results when they state: “When compared with placebo or oral morphine, participants gave lower pain intensity and higher pain relief scores for transmucosal fentanyl formulations at all time points”. First, the outcomes at 15 min (the most important to obtain a quick relief of pain), Pain intensity Difference (Comparison 2. Transmucosal opioid versus oral morphine) and Comparison 4. OTFC versus intravenous morphine), failed to show statistically significant differences with oral morphine (mean difference 0.37 CI 95% 0.00‐0.73) and with morphine iv (mean difference 0.80 CI 95% 0.00‐1,60). In any case the results had no clinical relevance. Moreover the authors state “at all time points”, whereas they do not provide any data for longer times (i.e. 30, 45, 60 min). In addition, authors state “transmucosal fentanyl citrate are safe (..) (compared with both placebo and morphine) in relieving breakthrough pain”. Surprisingly no analysis of adverse events have been done that were only described in each study. Seven out 15 were crossover trials and it was impossible to draw conclusions about it.
As reviewers we know that multiple comparisons could be made, but the most clinical interesting comparison is the gold standard, morphine.
The review only shows in SOF, comparisons of fentanyl with placebo and concludes, that it is effective. Surprisingly, there is no mention of morphine comparison, which we consider a key point, because no patient will take placebo if he has a breakthrough pain, but morphine for obvious reasons.
In our opinion, traversing the authors’ conclusions, this review did not show that the use of oral and nasal transmucosal fentanyl is an effective alternative to morphine for patients with breakthrough cancer pain.
Reply
The authors of the review were contacted but chose not to provide a response to the feedback.
Contributors
Kate Seers, PaPaS Feedback Editor, and Anna Hobson, PaPaS Managing Editor.
Additional feedback, May 2015
Summary
On 15 February 2015, Dr Vicente Ruiz Garcia (University Hospital La Fe, Spain) submitted feedback via the Cochrane Library. The main complaint is available in Feedback 1 above.
On 23 February, 16 March and 15 April, Kate Seers (Feedback Editor, PaPaS) contacted the authors and invited them to respond. On 21 March, John Zeppetella (lead author) declined to provide a formal response. No response was received from Andrew Davies (second author).
On 31 March, PaPaS sought advice from the Cochrane Editorial Unit (CEU) on how to manage the issue; advised to publish without a response, depending on nature of feedback.
On 11 May 2015, review re‐published with feedback incorporated.
On 20 May 2015, Marta Roqué Figuls (Statistician, Iberoamerican Cochrane Centre) wrote to the CEU repeating the initial claim. They did not agree with the approach decided upon by the Co‐Ed, which was supported by the EiC. They stated “The CCIb assessment is that the review presents methodological shortcomings, and the conclusions are skewed in favor of fentanyl. Consequently, we support Vicente and colleagues’ petition to re‐assess the publication status of the review.”
Reply
On 9 June, Christopher Eccleston (Co‐ordinating Editor, PaPaS) advised that the review remain unchanged until it was ready for updating in September 2015.
On 25 June 2015 David Tovey (Editor in Chief, Cochrane) responded to say “We have now had a chance to appraise this review and also obtain a report from our screening team… In summary we agree with almost all of the criticisms made of the review, and are concerned that the flaws may mean that the findings are misleading as currently presented. We note that the authors have declined to respond to the useful comments provided by Vicente. Having considered this and discussed it internally, we agree with our colleagues at the IbCC that the review should be withdrawn temporarily until the errors have been fixed and the review updated. We would like to re‐screen the review before any update or amendment is published.”
On 2 July and 23 July, Anna Hobson (Managing Editor, PaPaS) again invited the authors to respond to the initial feedback and subsequent reviews by 23 July. No response was forthcoming.
At August 2015, the review was withdrawn. See Published notes for more information.
Contributors
Kate Seers (Feedback Editor, PaPaS), Cochrane Editorial Unit (CEU), Christopher Eccleston (Co‐ordinating Editor, PaPaS), David Tovey (Editor in Chief, Cochrane), Anna Hobson (Managing Editor, PaPaS).
What's new
| Date | Event | Description |
|---|---|---|
| 12 October 2015 | Amended | Feedback now accessible in Published notes. |
History
Protocol first published: Issue 3, 2003 Review first published: Issue 1, 2006
| Date | Event | Description |
|---|---|---|
| 13 August 2015 | Amended | Additional feedback has been added, and this review has now been withdrawn. See Feedback 2 and Published notes. |
| 11 May 2015 | Feedback has been incorporated | Feedback on this review was submitted in February 2015. See Feedback section for details. |
| 11 February 2013 | New citation required but conclusions have not changed | None of the eleven additional studies identified significantly changed the outcome of the review. |
| 6 February 2013 | New search has been performed | Twenty one additional studies were identified in the updated search, eleven were included (Fallon 2011; Kress 2009; Lennernäs 2010; Mercadante 2007; Mercadante 2009; Portenoy 2006; Portenoy 2010; Rauch 2009; Rauch 2010; Rauch 2012; Slatkin 2007), and ten excluded (Ashburn 2011; Davies 2011; Hagen 2010; Johnson 2010; Rauch 2011; Rauch 2012b; Reynolds 2011; Reynolds 2012; Stanley 2011; Taylor 2010). |
| 7 January 2013 | Amended | Converted to new review format |
| 13 October 2008 | Amended | Converted to new review format. |
Sources of support
Internal sources
None, Other.
External sources
None, Other.
Withdrawn from publication for reasons stated in the review