Letter to the Editor Letters to the Editor will be published, if suitable, as space permits. They should not exceed 1000 words (typed double-spaced) in length and may be subject to editing or abridgment. Response to Resistance Training, Blood Pressure, and Meta-Analyses We thank Rossi et al 1 for their interest in our work 2 and appreciate the opportunity to answer their comments. 1 Our recent meta-analysis 2 aimed to provide an update of the previously published meta-analysis on the effect of resistance training (RT) on blood pressure (BP). 3 We included all of the randomized, controlled trials that reported the effect of RT on BP in healthy adults. By contrast, Rossi et al 1 suggested that we should only have included articles in which BP was the primary outcome. They referred to Dwan et al 4 to state that the inclusion of secondary outcomes might have influenced the effect estimates in our meta-analysis. Concordant to others, 5 Dwan et al 4 provide evidence for the fact that studies with statistically significant results, as well as statistically significant outcomes, are more likely to be published than studies that report no statistically significant results or nonsignificant outcomes. This leads to a bias in the published literature that can then carry over to a meta-analysis. This problem, known as publication bias, is not new, and we fully agree that it can affect the effect size. In hopes of minimizing the bias, the best approach is to perform a truly comprehensive search of the literature. 5 By including studies that did not have BP as primary outcome, we attempted to be as complete as possible. In addition, several methods have been developed to assess publication bias. One such method is the funnel plot. 5 For the effect of dynamic RT on systolic and diastolic BP, both funnel plots appeared to be reasonably symmetrical and did not suggest publication bias. 2 In addition, I 2 , which is a measure of inconsistency across the findings of the studies, was 35% for both, which can be interpreted as “might not be important,” and so we feel that it is reasonable to state that our findings are more or less robust. However, in response to Rossi et al, 1 we have reanalyzed our results including only studies that specifically reported that BP was one of the primary outcomes. Contrary to Rossi et al, 1 we could include 11 random- ized, controlled trials involving 13 study groups. A smaller but still significant reduction of systolic/diastolic BP of 2.7 mm Hg (95% confidence limit: -4.8 to -0.54 mm Hg)/1.9 mm Hg (95% confidence limit: -3.3 to -0.54 mm Hg) could be observed, and there was no difference between studies with BP as a primary or secondary outcome (P0.50). In addition, I 2 equals 0 for systolic and diastolic BP, indicating a very consistent result. Given that the meta-analysis of Rossi et al 1 was not published in a peer-reviewed journal, we regret not being able to compare the 2 meta-analyses. Next, we clearly define the eligibility criteria for inclusion, we describe all of the used databases with start and end date of the search, and we report the different key words used for our search. 2 Therefore, we do not agree with Rossi et al 1 that we did not follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Statement with regard to reporting of the search strategy and screening procedure used. However, we do agree that we did not show a flow diagram, mainly because our meta-analysis 2 is based on a database that was started in 1985 and updated in 1994, 1999, 2003, 2004, and again for the current meta-analysis. Hence, it was not possible to give a clear overview of the number of records retrieved, retained, and eliminated at the various stages. In conclusion, we believe that, given the lack of clear publication bias and the robustness of the observed findings, it is reasonable to suggest that dynamic RT has the potential to lower BP. But we agree with Rossi et al 1 that there is still a need for large randomized, controlled trials investigating the effect of RT on BP, particularly in the hypertensive patient. 2 Sources of Funding V.A.C. is supported as a postdoctoral fellow by Research Foundation Flanders (FWO). Disclosures None. Ve ´ronique A. Cornelissen Research Center for Cardiovascular and Respiratory Rehabilitation Department of Rehabilitation Sciences Faculty of Kinesiology and Rehabilitation Sciences KU Leuven Leuven, Belgium Robert H. Fagard Hypertension and Cardiovascular Rehabilitation Unit Department of Cardiovascular Diseases Faculty of Medicine KU Leuven Leuven, Belgium Luc Vanhees Research Center for Cardiovascular and Respiratory Rehabilitation Department of Rehabilitation Sciences Faculty of Kinesiology and Rehabilitation Sciences KU Leuven Leuven, Belgium 1. Rossi A, Moullec G, Lavoie KL, Bacon SL. Resistance training, blood pressure, and meta-analyses. Hypertension. 2012;59:e22– e23. 2. Cornelissen VA, Fagard RH, Coeckelberghs E, Vanhees L. Impact of resistance training on blood pressure and other cardiovascular risk factors: a meta-analysis of randomized, controlled trials. Hypertension. 2011;58: 950 –958. 3. Cornelissen VA, Fagard RH. Effect of resistance training on resting blood pressure: a meta-analysis of randomized controlled trials. J Hypertens. 2005;23:251–259. 4. Dwan K, Altman DG, Arnaiz JA, Bloom J, Chan AW, Cronin E, Decullier E, Easterbrook PJ, Von Elm E, Gamble C, Ghersi D, Ioannidis JP, Simes J, Williamson PR. Systematic review of the emperical evidence of study publication bias and outcome reporting bias. PloSone. 2008;3:e3081. 5. Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. Introduction to Meta-Analysis. 1st ed. The Atrium, Southern Gate, Chichester, West Sussex, United Kingdom: John Wiley & Sons Ltd. (Hypertension. 2012;59:e24.) © 2012 American Heart Association, Inc. Hypertension is available at http://hyper.ahajournals.org DOI: 10.1161/HYPERTENSIONAHA.111.188979 e24 by guest on February 8, 2016 http://hyper.ahajournals.org/ Downloaded from