LAUSR

Thank you for allowing us to respond to the letter by Cross et al. [1], which we have read with great interest and amazement, given some of the unfounded comments made. First, we must emphasize that the meta-analysis published by Alcaraz et al. [2] did not intend to analyse the effects of resisted sled training (RST) loading and training status on sprint performance. This meta-analysis performed an updated systematic review to determine the effectiveness of RST on sprint performance, and secondarily, to explain which variables produced a greater improvement in sprint performance in both the acceleration and maximum velocity phases. Secondly, Cross et al. [1] criticize the operational basis of our review, when in fact all recommendations in the PRISMA statement [3] were followed. We clearly formulated the question using systematic and explicit methods to identify, select, and critically appraise relevant research, and collected and analysed data from studies included in the review using appropriate statistical methodology [3]. Publications providing guidance on how to conduct a systematic review with meta-analysis [4–9] suggest that one of the reasons for conducting this type of research is to resolve conflicts around an issue arising from controversial findings in the literature. We reiterate that one of our secondary aims was to analyse outcomes when different loads are used. We also disagree with the theory proposed by Cross et al. [1] regarding use of “heavy load” RST. First, training with the optimal power load (considered by these authors to be associated with a 50% decrement in maximum velocity and approximately 80% of body mass [BM]) has not been shown to be more beneficial than training with lower loads (i.e. loads that produce a loss of 10% of maximum velocity). In fact, this same research group has recently published a study [10] that found greater improvements (small effect sizes) in sprinting (5, 10 and 15-m) with 10% of velocity loss versus 50% loss (only trivial effect sizes). Secondly, it has not been directly demonstrated that the optimum power load is achieved at 50% of the maximum velocity with the use of RST. These authors estimated this value using a radar device to calculate horizontal force indirectly [11]. Although this device has been validated with a treadmill, we are not aware of any study that has calculated the mechanical power output by directly measuring force and velocity in a complete sprint. Therefore, we must be very cautious about statements made in relation to these estimates. Cross et al. [1] also questioned the need for our metaanalysis given that a systematic review on the topic had been published 2 years previously [12]. In fact, both that review [12] and an earlier one by Hrysomallis [13] found that RST performance benefits over unresisted sprint training could not be conclusively demonstrated. Our meta-analysis is a justifiable attempt, using a different approach, to clarify whether benefits exist or not. Cross et al. [1] question our cut-off of 20% of BM loading parameters. However, this reflects the fact that the few studies that have found kinematic modifications with different loads concluded that significant changes were generally observed after applying 15% [14] or 20% of BM [15]. Furthermore, a cut-off at the point where RST begins not This reply request to the letter available at: https ://doi.org/10.1007/ s4027 9-018-0947-8.