We would like to thank de la Gala et al. for their interest in this clinical trial.
First, the authors raise a concern about differences between the groups in amount of fluids given and the depth of anesthesia. It is important for any trial in perioperative pulmonary medicine to set rules for volume management, hemodynamics, and plateau pressure during ventilation as potential confounders in the study protocol. We were well aware of this issue and therefore established a detailed protocol for the study centers with only very few protocol violations as reported. In this large trial with 460 patients, we trusted the power of randomization to eliminate differences in these covariates between the two study arms.1 As a consequence, we are not concerned about the issues raised by the authors of the letter. The same argumentation applies to the Bispectral Index. The protocol instructed investigators to achieve Bispectral Index values of 40 to 60.
Second, the authors are concerned that we subsume the effect of desflurane under volatiles in general. As described in the letter, differences in clinical effects of the different volatile anesthetics do exist and of course we acknowledge this. Also, different antioxidant properties of volatile anesthetics have been described, which at the same time can also be attributed to propofol depending on the formulation and addition of EDTA.2 However, it was not our intent to compare different volatile anesthetics. Based on previous findings by our group, we do indeed assume that all halogenated anesthetics with trifluorinated carbon (–CF3) groups provide protection.3,4 It remains an interesting question if our results could be replicated using sevoflurane instead of desflurane or a different propofol formulation.
Third, the authors address the issue of anesthetics as being lung protective. When discussing protection, it seems key that we clearly define the endpoint of such a protection. As already shown in several randomized controlled trials, the lung itself benefits from a volatile anesthesia, most often through attenuation of inflammatory processes, triggered by ventilation and/or surgical manipulation.5–7 Based on the result of one of these studies, where patients were shown to have less complications after thoracic surgery with volatile anesthetics,6 we hypothesized that there would be fewer complications overall for patients in the desflurane arm of our trial. Importantly, the trial of De Conno et al.6 included only 54 patients with surrogate markers as the primary and the clinical outcome as the secondary endpoint. With postoperative complications using a widely accepted classification system as the primary endpoint and an adequately powered approach, our hypothesis was not confirmed with our trial. In conclusion, lung protection provided by anesthetics does not lead to an overall better postoperative outcome. Our trial also highlights the importance of precisely defining study endpoints and proper interpretation. Therefore, we do not think this hypothesis should be reevaluated as proposed by the authors of the letter. In contrast, further trials are needed to test the hypothesis if volatile anesthetics improve complications in the entirely different clinical scenarios of major organ injury such as transplantation, which we did not address in this study.
The authors declare no competing interests.