Re: Discrepancies in autologous bone marrow stem cell trials and enhancement of ejection fraction (DAMASCENE): weighted regression and meta-analysis
In the April 29, 2014 edition of the British Medical Journal, Nowbar et al. published a review of meta-analysis of autologous bone marrow stem cell trials for heart disease (DAMASCENE). The conclusions that they draw are curious.
Scientists aspire to perfection but do not claim that their manuscripts are perfect. Every manuscript can be improved, its data reexamined, each table and figure rechecked yet again. The “discrepancy” issue may be important insofar as it highlights the range of errors observable for a given field of study. It would, therefore, have been most useful to know the distributions of discrepancies, differentiating between minor (e.g., editing errors and data inconsistencies) and major ones such as misrepresentations of data or research designs. Yet this either is not available in the DAMASCENE article or buried in complex appendices. Also, it is curious that no perspective is provided on how the error rates in cell therapy trials compare to those in other research fields. This information is available, yet no such report was cited to provide a helpful perspective.
There are other concerns. Their use of an ecological design injects the ecological fallacy into their conclusions, a well-known weakness of analyses based on populations rather than on individuals. While others who use this macro approach acknowledge the inadequacies of this methodology, Nowbar is silent. Neither do the authors admit that only approximately 16% of the variability in the LVEF effect size they attempted to elucidate is actually explained by their work. Perhaps most critically, the true relationship of interest is not that between discrepancy ridden writing and effect size, but between poor research conduct and effect size. DAMASCENE does not address whether mistakes in writing imply mistakes in research design and execution. Without this link, the DAMASCENE argument collapses.
Meta-analytic critiques offer important contributions to the literature enabling readers to review facts and determine value for themselves. However, just as DAMASCENE suggests that publications should accurately reflect the facts, critiques can and should be rebutted where appropriate. One such example is the C-Cure investigators’ work which Francis et. al. critiqued. Although responsibly rebutted, there is no admission of this in the DAMASCENE article. Finally, Nowbar et al. state, “The institution of the SCIPIO trial is recently reported to have requested that the publication be retracted.” However, there is no evidence suggesting that SCIPIO’s home institution (University of Louisville) ever made this request. This inclusion of unverified, unattributed, and false allegations in the mainstream medical literature is misleading.
As the DAMASCENE article suggests, it is important that discrepancies be brought to the attention of readers in order to allow for a fair and balanced review. It is equally important to apply this standard to publications that claim to do so.
Lem Moyé, M.D., Ph.D.
PI- Coordinating Center – Cardiovascular Cell Therapy Research Network
1. Nowbar AN, Mielewczik M, Karavassilis M, Dehbi HM, Shun-Shin MJ, Jones S, Howard JP, Cole GD, Francis DP; DAMASCENE writing group.Discrepancies in autologous bone marrow stem cell trials and enhancement of ejection fraction (DAMASCENE): weighted regression and meta-analysis.BMJ. 2014 Apr 28;348:g2688. doi: 10.1136/bmj.g2688.
2. Costantino G1, Casazza G, Cernuschi G, Solbiati M, Birocchi S, Ceriani E, Duca P, Montano N.Errors in medical literature: not a question of impact. Intern Emerg Med. 2013 Mar;8(2):157-60. doi: 10.1007/s11739-012-0880-z. Epub 2012 Nov 22.
3. Greenland S, Robins J (1994) Ecologic studies-biases, misconceptions, and counterexamples. Am J Epidemiol 139:747–760.
4. Yamori Y1, Liu L, Mizushima S, Ikeda K, Nara Y; CARDIAC Study Group.Male cardiovascular mortality and dietary markers in 25 population samples of 16 countries. J Hypertens. 2006 Aug;24(8):1499-505.d Ruth D. Peterson, eds
5. Bartunek J, Behfar A, Dolatabadi D, Vanderheyden M, Ostojic M, Dens J, El Nakadi B, Banovic M, Beleslin B, Vrolix M, Legrand V, Vrints C, Vanoverschelde JL, Crespo-Diaz R, Homsy C, Tendera M, Waldman S, Wijns W, Terzic A.Cardiopoietic stem cell therapy in heart failure: the C-CURE (Cardiopoietic stem Cell therapy in heart failURE) multicenter randomized trial with lineage-specified biologics. J Am Coll Cardiol. 2013 Jun 11;61(23):2329-38. doi: 10.1016/j.jacc.2013.02.071. Epub 2013 Apr 10. Erratum in: J Am Coll Cardiol. 2013 Dec 24;62(25):2457-8.
6. Mielewczik M, Cole GD, Nowbar AN, Schilling R, Whinnett ZI, Bordachar P, Wilmshurst P, Chambers JC, Olshansky B, Morgan J, Israel C, Sethi AS, van Houwelingen M, Cleland JG, Schmidt G, Francis DP. The C-CURE Randomized Clinical Trial (Cardiopoietic stem Cell therapy in heart failURE). J Am Coll Cardiol. 2013 Dec 24;62(25):2453. doi: 10.1016/j.jacc.2013.09.013.
7. Bartunek J, Behfar A, Dolatabadi D, Vanderheyden M, Ostojic M, Dens J, El Nakadi B, Banovic M, Beleslin B, Vrolix M, Legrand V, Vrints C, Vanoverschelde JL, Crespo-Diaz R, Homsy C, Tendera M, Waldman S, Wijns W, Terzic A. Reply: The C-CURE Randomized Clinical Trial (Cardiopoietic stem Cell therapy in heart failURE). J Am Coll Cardiol. 2013 Dec 24;62(25):2454-6. doi: 10.1016/j.jacc.2013.09.014.
Competing interests: No competing interests