LAUSR

We thank Aleksandra Goch et al. for showing interest in our literature and for providing additional comments regarding the limitations of our paper. Regarding the first limitation on calculated plasma volume status (PVS) value used in the first model, as she pointed out, the Optimized CathEter vAlvular iNtervention (OCEAN)transcatheter aortic valve implantation (TAVI) registry cohort is divided into quartiles as follows: PVS < 5.5%, 5.5% < PVS < 13.5%, 13.5% < PVS < 21.0%, PVS > 21.0%. However, this is not a cut-off value for PVS, but a simple quadrant of the OCEAN-TAVI registry cohort. The purpose of the first model was to demonstrate that the prognosis worsens with increasing PVS. In effect, the prognosis of post-TAVI patients worsened with increasing PVS. The best discriminated cut-off value of all-cause mortality was PVS 19.0% in the OCEAN-TAVI registry. We identified this cut-off value by using the survival classification and regression tree (CART) method. Conversely, Maznyczka et al. used Youden’s index for predicting the best cut-off value of PVS 0%. Youden’s index is an index based on the sum of sensitivity and specificity. If the sum of the sensitivity and specificity is high, the diagnostic performance seems adequate. However, the required magnitude of sensitivity and specificity varies depending on the purpose of the study; therefore, it seems a little rough to mechanically use a high Youden’s index as the optimal cut-off. In addition, the cut-off value of PVS for all-cause mortality varies greatly depending not only on the number of patients but also on race, patient size and background, and post-treatment medication and follow-up methods. As mentioned in our paper, the PVS of 19.0% is only a rough guide. It is also information obtained by using the specific old-age Asian cohort. In addition, the most important point is that an increase in PVS can be an LETTER TO THE EDITOR