Study On Prevention And Treatment Of Acute Respiratory Failure Following Extubation By Non-invasive Respiratory Support In Critical Ill Patients

BackgroundThe re-intubation rate of mechanical ventilation patients after extubation is about10%-20%,Extubation failure resulting in prolonged mechanical ventilation,increased frequency of ventilator-associated pneumonia,and longer intensive care unit（ICU）and hospital stays,increased mortality（25–50%）.Therefore,It is essential to identify patients at high risk of postextubation ARF in order to choose an appropriate strategy of respiratory support able to improve their outcome.Conventional oxygen therapy（COT）、Non-invasive ventilation（NIV）and high-flow nasal oxygen（HFNO）have been implemented as effective approaches aimed at protecting extubation.Although current clinical evidences have shown a lower incidence of acute respiratory failure after extubation in high-risk patients using NIV or HFNO compared with COT,the mechanism is unclear.Differences in inspiratory effort and pulmonary gas distribution may be important reference criteria for the selection of sequential respiratory support after extubation.In conclusion,we hypothesized that NIV or HFNO could reduce inspiratory effort and work of breathing,improve lung homogeneity in high-risk patients with extubation failure compared with COT.Acute respiratory failure（ARF）is the most common complication after major surgery,especially lung surgery.The mortality of patients undergoing invasive mechanical ventilation after traditional lobectomy with ARF is as high as 60-80%.Invasive mechanical ventilation is closely associated with pulmonary infection,persistent pleural leakage and bronchopleural fistula.Therefore,it is important to avoid endotracheal intubation and invasive ventilation when respiratory failure occurs after lobectomy.NIV has been reported to improve atelectasis and gas exchange,reduce complication,re-intubation rate and mortality,and shorten ICU stay in ARF compared with COT.However,there have been few clinical studies on NIV in the treatment of ARF after lobectomy,especially for patients with ARF after lobectomy with VIDEO-assisted Thoracoscopic Surgery（VATS）,and the effectiveness of NIV has not been verified.Therefore,we hypothesized that NIV is an effective respiratory support for ARF after lobectomy of VATS.Objectives1.Explore the effects of sequential COT,HFNO and NIV on gas exchange,respiratory mechanics and gas distribution after extubation.To study the mechanism of NIV and HFNO in preventing ARF after extubation.2.Observe the efficacy of NIV in the treatment of ARF after VATS lobectomy in the real world,and explore the high risk factors of NIV failure.Methods Part 1A single-center,prospective,randomized crossover design was used to screen patients admitted to the Department of Intensive Care Medicine of the First Affiliated Hospital of Guangzhou Medical University from January 2021 to March 2022 for more than 72 hours of ventilation,scheduled extubation and high risk factors for extubation failure.After extubation,patients were randomly and sequentially treated with COT,HFNO and NIV.Each respiratory support was maintained for 30min,and data including vital signs,blood gas analysis,visual analog score,esophageal pressure measurement,and EIT monitoring were recorded 5min before the end of each respiratory support.Gas exchange(p H,Pa O₂,Pa CO₂,Pa O_2/Fi O₂and ROX index,respiratory mechanics(RR,△P_es,transpulmonary pressure at the end of expiratory(PL_ee),transpulmonary pressure at the end of inspiratory(PL_ei),△P_L,PTP_pes),hemodynamics（MAP,HR）,gas distribution（GI index,△EELI,regional ventilation ratio（Z%ROI1,Z%ROI2,Z%ROI3,Z%ROI4,Z%ROI1+ROI2,Z%ROI3+ROI4））were compared among different respiratory support treatments.The differences of the above indicators were further analyzed and compared among subgroups.Part 2The clinical data of VATS lungectomy patients admitted to the Intensive Care Unit of the First Affiliated Hospital of Guangzhou Medical University from January 2019 to June 2021 were retrospectively analyzed.First,patients with postoperative ARF who received NIV was screened.Secondly,ARF patients were divided into hypercarbonate ARF group and non-hypercarbonate ARF group according to whether accompanied by carbon dioxide retention.The re-intubation rate,length of ICU and hospital stay,NIV time,ICU mortality,in-hospital mortality were compared between the two groups.Finally,ARF patients were divided into successful NIV group and NIV failure group according to whether they were eventually converted to invasive mechanical ventilation,and the risk factors for NIV failure were analyzed.Results Part 11.Baseline characteristics:A total of 20 patients at high risk of extubation failure were included,including 13 ARDS patients and 7 patients after lung transplantation,with an average age of 59.5（44-69）years old and male majority（16 cases,80%）,with an average APACHE II score of 19（14-24）and an average SOFA score of 9（7-11）.Before extubation,the average Pa O₂/Fi O₂ was 286（248-324）mm Hg,and the average Pa CO₂was 34.1（31.5-42.3）mm Hg.2.Gas exchange and hemodynamics:There were no statistically significant differences in p H,Pa O₂,Pa CO₂,Pa O₂/Fi O₂,ROX index,MAP and HR among COT,HFNO and NIV after extubation.3.VAS score:The score of dyspnea of HFNO was significantly lower than that of COT（2[1-2]vs.2[1-3],P<0.05）,and there was no statistical difference between NIV with COT or HFNC.The comfort scores of HFNO and COT were better than those of NIV（3[2-4]vs.2[2-2],P<0.01）（3[2-4]vs.2[1-2],P<0.01）,while there was no statistical difference in comfort scores between HFNO and NIV.4.Respiratory mechanics:Compared with COT,HFNO and NIV could increase P_Lee（-3.1±4.9 cm H₂O vs.-7±6.5 cm H₂O,P=0.026;-2.4±4.5 cm H₂O vs.-7±6.5 cm H₂O;P=0.09）,decreased PTP_Pes(58.6±21.5 cm H₂O*s/min vs.88.8±56.9 cm H₂O*s/min,P=0.017;NIV also decreased△P_es（5.2±2.2 cm H₂O vs.8.5±4.1 cm H₂O,P=0.002）.P=0.003),but had no significant effect on RR,end breath P_esand△P_L.Compared with HFNO,The△P_L of NIV was higher（10.2±4.1 cm H₂O vs.7±3.5 cm H₂O,P=0.013）,but there were no statistical differences in RR,△P_es,end breath P_es,P_Lee and PTP_Pesbetween the two groups.5.Gas distribution:GI index of HFNO and NIV was lower than COT（0.39±0.05 vs.0.47±0.1,P<0.01;0.38±0.08 vs.0.47±0.1,P<0.01）.The△EELI of HFNO and NIV was higher than that of COT（0.67±1.32 vs.-0.79±1.1,P<0.01;0.51±1.01 vs.-0.79±1.1,P<0.01）,the difference was statistically significant,and the difference was mainly reflected in the gravity dependent zone（3+4 zone）.There was no significant difference in GI and△EELI between HFNO and NIV.There were no significant differences in RVD,CW,CL and COV among the three groups.6.Prolonged intervention time:In 5 patients,COT,HFNO and NIV were extended to1h respectively,and there were no statistically significant differences in all gas exchange and hemodynamic indexes after 1h intervention with the three respiratory support methods.△P_esof COT was significantly higher than NIV（8.8±2.7 cm H₂O vs.4.8±1.4 cm H₂O,P=0.013）,but there was no statistical difference between NIV and HFNO（8.8±2.7 cm H₂O vs.6.4±2.2 cm H₂O,P=0.013）.P>0.05).In terms of PTP_pes,COT was significantly higher than HFNO group（70.4±29 cm H₂O*s/min vs.41.3±7.3 cm H₂O*s/min,P=0.032）and NIV group（70.4±29 cm H₂O*s/min vs.34.6±13.2cm H₂O*s/min,P=0.012）.In terms of gas distribution,The GI index of HFNO and NIV groups was lower than that of COT group（0.37±0.04 vs.0.42±0.04,P<0.05;0.34±0.02 vs.0.42±0.04,P<0.01）,the difference was statistically significant;The△EELI in HFNO and NIV groups was higher than that in COT group（0.18±0.54 vs.-0.92±0.50,P<0.05;0.49±0.79 vs.-0.92±0.50,P<0.01）,the difference was statistically significant.7.Subgroup population:For ARDS subgroup,GI index of HFNO and NIV was lower than COT（0.38±0.04 vs.0.47±0.1,P<0.05;0.38±0.09 vs.0.47±0.1,P<0.05）and△EELI were higher than COT（0.96±1.59 vs.-0.69±1.18,P<0.01;0.69±1.18vs.-0.79±1.1,P<0.05）.There was no statistical difference in GI index and△EELI between HFNO and NIV.For lung transplantation subgroup,HFNO and NIV had higher△EELI than COT（0.21±0.59 vs.-1.1±0.73,P<0.05;0.22±0.81 vs.-1.1±0.73,P<0.05）,the difference was statistically significant.There was no statistical difference in GI index among the three groups.Part 21.Baseline characteristics:A total of 1549 patients with VATS lobectomy were enrolled,of whom 67（4.3%）developed ARF and received NIV,including 15 patients（22.3%）with hypercarbonate ARF and 52 patients（77.7%）with non-hypercarbonate ARF,with a reintubation rate of 19.4%（13/67）.2.NIV efficacy:The overall re-intubation rate of ARF after NIV treatment of VATS was 19.4%（13/67）,and no patient died.There were no statistically significant in Re-intubation rate of non-hypercarbonate ARF group and hypercarbonate ARF group[19.2%（10/52）vs.20%（3/15）,P=0.95],NIV time[48（20-76）h vs.41（14-69）h,P=0.527],ICU duration[3（1-6）d vs.4（2-10）D,P=0.21]and hospitalization duration[13（7-21）vs.15（7-20）,P=0.986]between the two groups.3.High-risk factors of NIV failure:The respiratory rate of NIV failure group（30±5times/min vs.24±7 times/min,P=0.005）and the proportion of lymph node dissection[84.6%（11/13）vs.50%（27/54）,P=0.02]were higher than those of NIV success group,and the differences were statistically significant.Multivariate logistic regression analysis showed that high respiratory rate was an independent risk factor for NIV failure（OR 1.147,95%CI 1.012-1.3,P=0.032）.Conclusion Part 11.NIV and HFNO can prevent ARF after extubation in high-risk patients with extubation failure by improving alveolar collapse,reducing work of breathing,increasing end-expiratory lung volume and improving lung homogeneity.2.NIV reduced inspiratory effort more than HFNO,but HFNO is best tolerated by patients.3.Sequential NIV or HFNO support was superior to COT after extubation in ARDS and lung transplantation patients.Part 2NIV was effective in the treatment of ARF after VATS lobectomy,and higher respiratory rate was an independent risk factor for NIV failure.