The primary aim of this study is to assess the efficacy of post-operatory HFNC in reducing the incidence of hypoxemia after gynecological oncology surgery, compared to the standard application of O2 through a Venturi mask; The secondary objectives are to investigate the occurrence and entity of lung atelectasis, to evaluate diaphragmatic function and respiratory discomfort, and to evaluate the incidence of respiratory complications after seven days in the two groups.

Patients will be randomized into two groups: HFNC and Control. The patients will be studied with preoperative lung and diaphragmatic ultrasound. Standard general anesthesia will be administered in the two groups. Ultrasound will be performed at arrival in the recovery room (RR) and before discharge from the RR. In the HFNC group, high-flow O2 will be administered; in the control group standard O2 therapy with Venturi mask will be administered.

Arterial blood gas analysis upon arrival in the RR and after two hours of O2 therapy in both groups will be checked.

The incidence of post-operative respiratory complications will be monitored in the seven days following surgery.

Post-operative hypoxemia presents in 10-50% of patients undergoing surgery, and it is associated to unfavourable prognosis. Anesthesia-induced paralysis, the use of high oxygen concentrations (O2) and inadequate level of positive end expiratory pressure (PEEP), capnoperitoneum, Trendelenburg position, prolonged duration of surgery, contribute to pulmonary atelectasis, which are the principal cause of postoperative hypoxemia.

The use of high flow nasal cannulas (HFNC) has been recently studied in the treatment of hypoxemic patients in intensive care unit and in the immediate post-operatory period after abdominal and cardio-thoracic surgery. The role of post-operative HFNC in patients undergoing major gynecological oncology surgery has yet to be assessed.

Lung ultrasound is a safe and accurate bedside tool that allows to investigate pulmonary aeration and atelectasis developement. Diaphragmatic ultrasound is a recent bedside technique useful to assess diaphragm performance.

The principal aim of this study is to assess the efficacy of post-operatory HFNC in reducing the incidence of hypoxemia after gynecological oncology surgery, compared to the standard application of O2 through the Venturi mask. The secondary objectives are to investigate the occurrence and entity of lung atelectasis, to evaluate diaphragmatic function and respiratory discomfort, and to evaluate the incidence of respiratory complications after seven days in the two groups.

Study Design and Length:

this is an interventional, prospective, randomized (with standard and experimental groups) monocentric study. The study will be 12-months in length and it will be carried out in the operating rooms of the 6° floor of the IRCCS Fondazione Policlinico Universitario Agostino Gemelli.

Materials and Methods:

The patients eligible for recruitment will be randomized with a computerized system and will be included in wither the HFNC or C group.

Upon arrival in the operating room lung and diaphragmatic ultrasound will be performed according to the international recommendations of Point of Care. A SonoSite ultrasound with Convex probe, with abdominal presetting and depth set at 9-12 cm, will be used. The ultrasound probe will be positioned between the ribs with the marker facing cranially and position perpendicular to the thoracic cage, oriented along the longitudinal axis of the patient. The image generated this way will show the superior rib on the left the screen and the inferior rib on the right of the screen, both represented by acoustic shadowing. The pleural line will appear between the costal shadows as a slightly curved and eco-reflective line. In the healthy lung, it is possible to appreciate pleural "sliding" generated by the movements of the visceral pleura against the parietal pleura during each breath. This movement will be more pronounced along the lung bases and more blurred at the apices. Horizontal lines at regular intervals below the pleura that may be appreciated represent reverberation artefacts that are defined as "A lines." On the other hand, the "B lines," that are also know as "comets" are artefacts produced by the interface of the normally aerated lung parenchyma with higher-density areas. These will present as vertical lines that start from the pleura and develop up to the end of the image, covering the A lines and consensual to respiratory movements. Twelve images will be acquired, one for each thoracic section obtained through the following method: for each hemithorax, anterior, posterior and lateral portions will be recognized relative to the anterior and posterior axillary lines, respectively. Each section will be further subdivided into superior and inferior by the intermammillary line. The posterior images will be acquired with the patient lying in lateral decubitus. A score ranging from 0 to 3 according to Monastesse criteria will be assigned to each of the 12 quadrants. For each quadrant the worst (higher) score will be considered, indicating more severe aeration loss. Individual scores will be added to calculate the cumulative lung ultrasound (LUS) score (0-36).

0: Normal aeration = visualization of pleural sliding and A lines parallel to the pleura with < 3B lines.1: Mild loss of aeration = ≥3 B lines or multiple subpleural consolidations separated by normal pleura. 2: Moderate loss of aeration = multiple and coalescent B lines of multiple subpleural consolidations separated by thickened or irregular pleura. 3. Severe loss of aeration = parenchymal consolidations or multiple subpleural consolidations greater than 1×2 cm in dimensions.

For diaphragmatic ultrasound a linear high frequency probe will be used. The operator identifies the caudal border of the costophrenic sinus as the zone of transition from the artefactual representation of the aerated lung to the visualization of the diaphragm and liver on the right side. The diaphragm will be identified by the two hyperechoic lines of pleural and peritoneal membranes. In M-mode, both the respiratory phases are represented in the same frame. The diaphragm thickness at end-inspiration (TEI), at end-expiration (TEE) will be measured and the thickening fraction (TF) will be calculated using the formula TF = (TEI - TEE)/TEE.

The clinical and respiratory parameters from the patients' medical records will also be reported.

A venous access will be placed, according to normal clinical practice. The use of neuro-axial blocks (spinal or epidural), as well as the placement of a central venous catheter, arterial catheter, a second large bore venous access, will be left to the judgement of the anesthesist in charge of the patient. The induction of general anesthesia will follow a preoxygenation with inspired oxygen fraction (FiO2) =1 for three minutes. The intraoperative mechanical ventilation settings will be standardized, with tidal volume set at 8ml/kg of predicted body weight, PEEP 5 cmH2O, respiratory frequency set to maintain PaCO2 within physiological limits, I:E ratio of 1:2. In case of intraoperative arterial desaturation (SpO2 < 95%) different techniques will be adopted in the following order: recruitment maneuvers, increase in PEEP, and increase in FiO2.

After awakening from general anesthesia the patients will be moved to the RR where standard multiparametric monitoring will be started, an arterial blood sample will be drawn for blood gas analysis and a second lung and diaphragmatic ultrasound will be performed following the same criteria already listed above. Respiratory rate and a Visual Analogue Scale (VAS) for discomfort and dyspnoea will be filled out.

At this point, the patients selected for the C group will undergo conventional O2-therapy Venturi mask. In the patients belonging to the HFNC group, high flow O2 (60 lt/min) will be administered through the use of nasal cannulas with the Opti-Flow circuit. In both groups the starting FiO2 will be 21% and will be increased of necessary to obtain a SpO2>92. After two hours of therapy a new arterial blood gas analysis, a third lung-diaphragmatic ultrasound, respiratory rate, a Visual Analogue Scale (VAS) for discomfort and dyspnoea will be noted, before the patients are transferred to the ward.

On the seventh day after surgery all of the eventual episodes that may be classified as postoperative pulmonary complications (PPC) will be recorded (hypoxemia, pleural effusion, lung edema, pulmonary embolism, pneumonia, acute respiratory distress syndrome (ARDS), bronchospasm, lung aspiration, pneumothorax).

Population of the study

Statistical Considerations Sample Dimension The literature indicates that roughly 50% of patients, treated with O2 therapy, develops hypoxemia one hour from extubation. Assuming that in the experimental branch (arm) that percentage is 15%, considering a strength of 90% and an alpha =0.05, a sample size of 66 patients will be obtained (33 per branch), at which a 20% of dropout will be added to reach the final sample size of N = 80 patients (40 per arm).

Data Analysis The sample will be described in its clinical and demographic characteristics through the techniques of descriptive statistics. In particular, quantitative variables will be represented though the following measurements: minimum, maximus, range, mean and standard deviation, or median and interquartile range. Qualitative variables will be represented through absolute frequencies and percentages. The normality of continuous variables will be validated through the Kolmogorov-Smirnov test. The primary aim will be reached comparing the two proportions through the Chi-squared test, while the change in Pa/FiO2 at the different time points, among the two groups of patients will be assessed by the Student's t-test. The secondary objective of comparing the incidence of PPC after seven days in the two branches will be also attained through the Chi-squared test. The p-value<0.05 indicates statistical significance. All of the statistical analysis will be done with SPSS 25.

Inclusion Criteria:

• patients scheduled for major gynecologic oncologic surgery (surgery duration > 2 hours).

Exclusion Criteria:

• patient's refusal to participate to the study
• pregnancy
• mechanical ventilation in the 2 weeks before surgery
• BMI>35