Erratum: Reply to Dr. Beltrami(J Appl Physiol DOI: 10.1152/JAPPLPHYSIOL.00822.2020er)

TO THE EDITOR: We appreciate the opportunity to respond to Dr. Beltrami's critical comments in his Letter (1) about our recent publication (3). Below, we paraphrase each of the main criticisms and follow each with our response. In his Letter, Dr. Beltrami writes the following: "First, maximal inspiratory pressure (MIP) . . .seems to have improved similarly in the placebo (85.6 vs. 101.2 cmH2O) and intervention groups (82.6 vs. 116.5 cmH2O, both P < 0.01 within groups) . . . . This issue alone precludes inferring causal relationship between MIP gains and vascular responses"(1). This is incorrect; the placebo group increased MIP by 19%, compared with 41% in the training group. However, we are in agreement with Dr. Beltrami: there is no "causal"link between MIP and blood pressure (BP) responses, and therefore, any improvement in MIP is immaterial. As stated in the INTRODUCTION and elsewhere, the purpose of the study is to determine the effects of inspiratory muscle strength training (IMST) versus sham control training on BP in this cohort, not to determine the relation between IMST-versus sham control- induced increases in MIP on BP. Dr. Beltrami also notes the following: "Second, the device used for assessing ambulatory 24-h BP. . .has only been validated against direct. . .auscultatory assessments, and it could be questioned the extent to which the values hold true. . . ."(1). Certainly, there are limitations to any form of BP measurement. However, ambulatory BP provides important insight into BP regulation in the context of life outside the laboratory. Moreover, any systematic errors introduced by the method itself would be comparable (magnitude and direction) in both the IMST and sham control groups. As such, the device used could not explain the different BP responses to IMST versus sham control training. In addition, Dr. Beltrami writes the following: "Third, the captions of Figs. 2 and 3 indicate that only a main effect of time was noted for laboratory and overnight BP, and results for the interactions are not given. . . ..[A]t no point do the authors compare the responses of the different groups"(1). Individual data points are presented for each of the main outcomes. Therefore, the changes in both groups are evident and easy to interpret. Interactions were assessed and found not significant, with one exception. A significant interaction was found for in-laboratory measures of casual systolic BP (P = 0.002). For this outcome, assessment of pre-post training effects clearly demonstrates a decline in systolic BP in the IMST group, whereas BP in the sham control group was unchanged. In concluding, Dr. Beltrami notes the following: ". . .important points of disagreement to the proposed theory have been left out, upselling the effects of IMT of blood pressure"(1). A comprehensive review of the IMST literature yields more than 35 studies, including studies in healthy young adults (4) and clinical populations (2). A critical consideration, however, is which among the studies includes an intervention that approximates the high-intensity, low-volume training adopted by Ramos-Barrera et al. (3). Although we have no wish to ignore the work of others, we are cognizant of important methodologic differences between these previously published training formats and the unique training stimulus of high-resistance, low-volume IMST used by Ramos-Barrera et al. (3). The data are clear, and we disagree that we are "upselling"the role of this type of IMST on blood pressure.