Prone Positioning in COVID-19 ARDS: Reply

Abstract

We wish to thank Dr. Zhao and Dr. Frerichs1 for acknowledging the clinical importance of the findings of our study.2 We agree with them that the correct implementation and, to a lesser extent, the precise methodology description of the electrical impedance tomography technique and analysis procedures are important for further development and clinical use.We accept that referring to “ventilated and perfused pixels” as the denominator in the calculation of shunt and dead space areas may be misleading and we should have clarified that the detected lung size was defined by all pixels ventilated and/or perfused. Nonetheless, we reassure Dr. Zhao and Dr. Frerichs that our calculation was performed correctly, and our mistake was a mere typo rather than a methodologic error.We also agree that the terms “dorsal ventilation” and “dorsal perfusion” are not in agreement with the consensus electrical impedance tomography terminology and definitions,3 as published in 2017. It is worth remarking, however, that we referred to the terms used in a more recent article,4 published in 2021, whose senior author was one of the experts participating in the consensus statement mentioned above. That said, we should probably have better defined our measurements as “dorsal ventilation area” (or “region” or “size”), indicating that we referred just to the pixel counts and not to the pixel values. Indeed, we stated in the Methods description that “dorsal ventilation represented the percentage of total ventilated lung area that is located in the dorsal half of the thorax.”The software kindly provided by Draeger for research purposes included the “homogeneity” rather than “inhomogeneity” index, and we accordingly used that parameter. We are led to believe this criticism should be directed to the company rather than to us. Nevertheless, we believe that the real issue about the inhomogeneity index is the strong dependency on the lung area considered for calculation. In the year 2008, when the inhomogeneity index was first introduced by Dr. Zhao,5 it was probably not considered that some conditions creating large “out-of-phase” variations—such as artifacts due to the heart and the diaphragm or to pleural effusions—would result in negative pixel values at end inspiration. Incorporating those out-of-phase pixels in the calculation of the global inhomogeneity index would lead to inclusion of extrapulmonary areas. Last, if pixels with negative values are excluded or set to 0, the inhomogeneity index cannot be greater than one.We agree that the assessment of lung perfusion with saline bolus has been studied in several animal studies; however, a comprehensive review on pulmonary perfusion with electrical impedance tomography was beyond the aims of our study. In all honesty, however, this limitation was not included in the original version of our article and was strongly and repeatedly required by one reviewer. Finally, while we may agree that an electrical impedance tomography image does not originate from “only the area of the lung surrounded by the belt,” it remains true that it does not cover the whole lung parenchyma, which makes the limitation still valid.Dr. Navalesi receives royalties from Intersurgical SPA (Mirandola, Italy) for the invention of Helmet Next. He also received speaking fees from Draeger (Lubeck, Germany), Intersurgical SPA, Getinge (Cinisello Balsamo, Italy), MSD (Rahway, New Jersey), Gilead (Foster City, California), and Novartis (Basilea, Switzerland). His research lab received research grants and/or research equipment from Intersurgical SPA, Draeger, and Gilead. Dr. Zarantonello and Dr. Sella received speaking fees from Getinge. The other authors declare no competing interests.