Reason for withdrawal from publication
This review has been withdrawn as authors are unable to complete the updating process.
The editorial group responsible for this previously published document have withdrawn it from publication.
Feedback
Clarification of definitions, 18 June 2013
Summary
Thank you for your review on preoperative statin therapy for patients undergoing cardiac surgery. We found it was thorough in light of the limited evidence in this area. We appreciate your decision to evaluate mortality as the primary outcome. However, it is unclear if attempts were made in communicating with study authors to consider whether death occurred even when it was not reported in the article. Second, it is unclear from the article whether participants had received statins before randomisation. It is not addressed whether any attempts were made in contacting study authors to identify whether participants received statins before randomisation. Five of the 11 articles included in the systematic review made no mention of the use of statins before randomisation in the study,1,2,3,4,5 which is fundamental to properly illustrate the direction of outcomes in the systematic review. If patients have been taking statins for several years before randomisation, a single dose of a statin may or may not yield additional benefit. Conversely, stopping the medication before the operation with a placebo arm may have a different effect. Third, studies provided widely varied definitions or lacked a definition for atrial fibrillation. We identified three articles that provided a clear definition of post‐surgery atrial fibrillation (i.e. as an episode captured by a monitor/12‐lead ECG lasting longer than five minutes with or without symptoms).6,7,8 However, the remaining studies did not define atrial fibrillation. With the current data given, we calculated an absolute risk reduction of 17% and a number needed to treat for an additional beneficial outcome of 5. Given the missing definitions, however, it is difficult to determine whether this would constitute a clinically relevant episode of atrial fibrillation. For example, one study may define atrial fibrillation as any asymptomatic episode documented on the ECG, and another as a symptomatic episode resulting in haemodynamic instability requiring intervention. We hope that the authors of this review will provide clarification of these points, and we look forward to hearing from you soon.
Sincerely, Luo (Lora) Wang, BSc Pharm Krystin Boyce, BSc, BSc Pharm Lawrence Nichoe Huan, BSc Pharm Anthony Amadio, BSc Pharm, ACPR Aaron M Tejani, BSc Pharm, PharmD
Reply
Dear Mrs. Wang,
We would like to thank you for your feedback on our systematic review and are grateful for your important questions. We hope that we have addressed all aspects properly.
Comment 1: “We appreciate your decision to evaluate mortality as the primary outcome. However, it is unclear if attempts were made in communicating with study authors to consider whether death occurred even when it was not reported in the article.”
Answer 1: Outcome variables of included studies were entered into the analysis as they were reported. In case of non‐reporting of an endpoint of interest, this variable was judged as “not assessed”, and no further contact was made with study authors. In general, study authors were contacted for various reasons. First, authors of eligible studies were asked whether or not outcome variables of interest were assessed. Specific studies were included or excluded accordingly. Second, definitions of endpoints need to be clarified in some cases to enhance comparability among studies.
Comment 2: “Second, it is unclear from the article whether participants had received statins before randomisation. It is unaddressed whether any attempts were made in contacting study authors to identify whether participants received statins before randomisation. Five of the 11 articles included in the systematic review made no mention of the use of statins before randomisation in the study,1,2,3,4,5 which is fundamental to properly illustrate the direction of outcomes in the systematic review. If patients have been taking statins for several years before randomisation, a single dose of a statin may or may not yield additional benefit. Conversely, stopping the medication before the operation with a placebo arm may have a different effect.” Answer 2: Eleven randomised controlled trials were included in this systematic review about statin therapy given before cardiac surgery. All included studies had differences concerning the duration of preoperative intake of the study medication and the proportion of participants taking statins before they were enrolled into the study. We totally agree that information on long‐term statin therapy for all included patients would have improved the clarity of the findings of this systematic review; however, given the small number of studies reporting on statin medication used before study entry, we did not perform a subgroup analysis of data from these studies. Unfortunately, we did not receive additional information on long‐term use of statin medications when we contacted the study authors.
Furthermore, we are currently investigating the impact of acute statin loading in patients undergoing coronary artery revascularisation in a multi‐centre randomised controlled trial (StaRT CABG trial; www.start‐cabg.de). Participants given a long‐term statin regimen for at least 30 days receive high‐dose statin therapy before undergoing the procedure with the aim of recapturing beneficial statin effects.
Comment 3: “Third, studies provided widely varied definitions or lacked a definition of atrial fibrillation. We identified three articles that had a clear definition of post‐surgery atrial fibrillation (i.e. as an episode captured by a monitor/12‐lead ECG lasting longer than five minutes with or without symptoms).6,7,8 However, the remaining studies did not define atrial fibrillation. With the current data given, we calculated an absolute risk reduction of 17% and a number needed to treat for an additional beneficial outcome of 5. Given the missing definitions, however, it is difficult to determine whether this would constitute a clinically relevant episode of atrial fibrillation. For example, one study may define atrial fibrillation as any asymptomatic episode documented on the ECG, and another as a symptomatic episode resulting in haemodynamic instability requiring intervention.”
Answer 3: We totally agree that variations in the definition of study endpoints present challenges for interpretation of results. As we already highlighted in the Methods section of our systematic review, the definitions of myocardial infarction, atrial fibrillation and renal failure show important discrepancies. Therefore, we accepted the definitions of these outcome variables as applied by the included trials.
We would be happy to answer further questions and thank you again for your thoughtful comments.
With best regards,
Oliver J. Liakopoulos and Elmar W. Kuhn
Contributors
Luo (Lora) Wang, BSc Pharm Krystin Boyce, BSc, BSc Pharm Lawrence Nichoe Huan, BSc Pharm Anthony Amadio, BSc Pharm, ACPR Aaron M Tejani, BSc Pharm, PharmD
What's new
| Date | Event | Description |
|---|---|---|
| 1 August 2017 | Amended | This review has been withdrawn as the authors are unable to complete the updating process. |
History
Protocol first published: Issue 4, 2010 Review first published: Issue 4, 2012
| Date | Event | Description |
|---|---|---|
| 19 May 2016 | Amended | The authors are currently updating this review as new evidence is available which has the potential to change the conclusions of this review. In the meantime, this review is withdrawn. |
| 23 June 2014 | New search has been performed | This version is an update |
| 23 June 2014 | New citation required and conclusions have changed | In this update, 6 studies were added to the review, resulting in a doubled number of participants. Of all 17 included studies, only 2 reported on the primary outcome (mortality) |
| 23 July 2013 | Feedback has been incorporated | Feedback incorporated on clarification of 3 points: contact with all study authors to obtain information on mortality in studies, identification of which participants received statins before randomisation and definition of atrial fibrillation |
Sources of support
Internal sources
No sources of support., Other.
External sources
No sources of support supplied
Withdrawn from publication for reasons stated in the review