Extubate Before Venovenous Extracorporeal Membranous Oxygenation Decannulation or Decannulate While Remaining on the Ventilator? The EuroELSO 2019 Weaning Survey

Abstract

Patients who receive venovenous extracorporeal membrane oxygenation (VV ECMO) support upon failure of invasive mechanical ventilation (IMV) and other adjuncts, including low tidal volume, prone positioning, adjusted positive end-expiratory pressure, and lung recruitment maneuvers, typically remain on IMV after VV ECMO commencement.1,2 Once both IMV and VV ECMO support are initiated, it is unclear which modality should be weaned first. Moreover, the optimal IMV settings, risk versus benefits of early spontaneous breathing, and optimal timing of VV ECMO weaning practices are yet to be defined. In addition, although the practice of liberation from IMV during ECMO3–8 has been increasing, the data concerning the weaning processes from ECMO and IMV are limited.9,10 Differences in weaning strategies for VV ECMO are based on the preferable approach is whether to wean ECMO or not when the patient is still mechanically ventilated.11 The approach to enable spontaneous breathing and early IMV weaning may prevent the complications of sedation allowing to avoid the risk of ventilation-induced lung injury and ventilator-associated pneumonia.5,6,12,13 A pragmatic way to “wean” as soon as the tidal volume start to recover has been proposed by clinicians at the Karolinska Institute ECMO Centre in Stockholm.11 Vasques et al. proposed a physiology-based assessment protocol, which combines an objective assessment of the native and artificial lung function. This method quantifies the patient’s response to a standardized weaning trial.9 However this strategy has not been tested in prospective trials and carries some challenges and risks in managing awake patients on ECMO. The aim of this survey was to understand current VV ECMO and IMV weaning practices globally. Materials and Methods This mixed-method online survey was conducted via SurveyMonkey14 from April 11, to May 31, 2019. It was distributed via three ways: the website of 8th EuroELSO15 Congress 2019 in Barcelona, the newsletter of worldwide Extracorporeal life Support Organization (ELSO),16 and social media of ELSO and EuroELSO (i.e., Facebook, LinkedIn, Twitter). The survey was developed by the authors through an interactive approach according to the CHERRIES statement.17,18 It consisted of 15 open-ended and multiple-choice questions with no requirement to answer all questions (Table 1). The responders were asked to only consider patients receiving VV ECMO for respiratory failure. Participation was voluntary. More than one response was allowed from a center. In accordance with General Data Protection Regulation, no respondent data was stored. Institutional Review Board review was considered to be exempt prior the data such that individual subjects cannot be identified in any way. Table 1. - Survey Questionnaire and Results for Questions 1–10 Question Answers Number Percentage 1 Is you center an ELSO member centre?Responded: 253 Skipped: 0 a. Yes 170 67% b. No 73 29% c. I don’t know 10 4% ELSO number if applicable 2 Our ECMO center is a (mark one)Responded: 253 a. Respiratory ECMO centre 25 10% b. Cardiac ECMO centre 5 2% c. Both respiratory and cardiac ECMO centre 223 88% 3 Which age groups does your center treat? a. Neonatal only 2 0.8% Neonatal, 0–30 days b. Pediatric only 1 0.4% Pediatric, 1 month to 16–18 years* c. Adult only 118 47% Adult, >16–18 years* d. Neonatal and paediatric 27 10.4% *Depending on local policy or legislationResponded: 253 e. Pediatric and adult 24 9.4% f. Neonatal, pediatric and adult 81 32% 4 How many annual runs does your center perform?Responded: 253 a. 0–5 19 7% b. 6–14 55 22% c. 15–30 56 22% d. >30 123 49% 5 Is tracheostomy usually performed in patients during the ECMO run? Responded: 253 a. Yes 122 48% b. No 131 52% 6 If yes, how many days after intubation?Responded: 118 Skipped: 135 a. Less than 7 days 21 18% b. 7–10 days 41 35% c. 10–14 days 45 38% d. 15–21 days 11 9% e. More than 21 days 0 0 7 When does your center extubate your patients when awake? (You can select more than one answer) a. During the ECMO run, before ECMO decannulation 61 50% ** b. After the ECMO decannulation 92 75% ** Responded: 122 Answers in total: 168 Skipped: 131 **Percentage of responders not answers c. ECMO patients on pulmonary support are not intubated staying awake during complete ECMO run 15 12% ** 8 If your center does extubate patients while on ECMO, how do you estimate what percentage of your total number of ECMO patients is extubated while on ECMO?Responded: 253 a. <10% 187 74% b. 10%–30% 37 15% c. 30%–50% 12 5% d. >50% 14 5% e. All 3 1% 9 Is the use of sedation at your institution score (RASS, Ramsay, others) targeted? Responded: 253 a. Yes 229 90.5% b. No 24 9.5% 10 The bedside nurse to patient ratio in ECMO patients at your institution is: (Please mark one answer)Responded: 253 a. 2 or more bedside nurses: 1 patient 46 18% b. 1: 1 143 56.5% c. 1: 2 45 18% d. 1: 3 or more patients 19 7.5% ELSO, Extracorporeal life Support Organization; ECMO, venovenous extracorporeal membrane oxygenation. Results Of 253 responses, the majority (67%) were from an ELSO-member center. ECMO was offered for both pulmonary and cardiac support in 88% of the centers and for pulmonary in 10%, for cardiac in 2%. Almost half (47%) of the centers treated only adults, and 32% all age-groups (neonatal, pediatric, and adults). Nearly half (49%) of the centers were reported to perform >30 ECMO runs per year. Over half (56.5%) of the centers provided a bedside nurse-to-patient ratio of 1:1. Ninety percent of participants titrated and monitored sedation according to a score (e.g., RASS, Ramsay), and 48% performed tracheostomy during the ECMO run, typically days 7–14 after intubation. Extubation was carried further decannulation from ECMO by 75% of respondents, whereas 12% aimed for awake and extubated patients during ECMO. Less than half (42%) claimed to reduce sedation and wean IMV to spontaneous ventilation during ECMO and 32% of centers to reduce sedation after commencement of ECMO, with a goal of liberating from IMV before ECMO decannulation (Table 2). Approximately half of the participants reported that their strategy depends on the underlying disease of the patient. Early extubation is considered more commonly for patients with chronic obstructive respiratory deisease or those awaiting lung transplantation than for those with adult (or acute) respiratory distress syndrome (ARDS) (Q12) (Table 2). Although 82 (32%) participants reported they prefer to liberate from IMV during ECMO answering question 11, only 15 participants (12%) reported that their patients are extubated while on ECMO (Q7). Thirty-two percent expressed concerns that awake and extubated patients are difficult to handle while on ECMO (Table 2). One out of four responders made amendments in free text to their multiple-choice answers with additional information of which some included additional reasons for the choice of the weaning strategy (Q11–13). Table 2. - Survey Questionnaire and Results for Questions 11–13 Question Answers Number Percentage 11 How would you describe the weaning strategy in ARDS patients on VV ECMO? (Please mark one)Responded: 253 Skipped: 0 a. Sedation after ECMO cannulation, controlled mechanical ventilation during ECMO, extubation or tracheotomy after ECMO decannulation, the patient is liberated / separated from ventilator after ECMO decannulation 41 16% b. Sedation after ECMO cannulation, gradually reduced sedation, assisted spontaneous breathing is allowed and implemented during ECMO, extubation or tracheotomy after ECMO decannulation, the patient is liberated / separated from ventilator after ECMO decannulation 105 42% c. Sedation at the beginning of ECMO; after ECMO cannulation, gradually reduced sedation, assisted spontaneous breathing is allowed and implemented during ECMO, extubation or tracheotomy during ECMO, patient is liberated / separated from ventilator before ECMO decannulation, awake ECMO weaning 82 32% d. Our center performs different weaning strategies in ARDS patients on VV ECMO (please describe) 25 10% 12 Please indicate whether the weaning strategy at your center depends on the reason for respiratory failure (You can select more than one answer)Responded 253 Answers in total: 337 Skipped: 0 **Percentage of responders not answers a. Yes, we tend to avoid intubation or early extubate patients awaiting lung transplant on ECMO 64 25% ** b. Yes, we tend to avoid intubation or early extubate patients with acute exacerbation of COPD 44 17% ** c. Yes, we tend to discontinue ECMO before attempting extubation in patients with ARDS 140 55% ** d. No, our weaning strategy for patients on venoveous ECMO is grossly the same regardless of the reason for respiratory failure 78 31% ** e. Yes, other (please specify) 11 4% ** 13 We would like to understand the reason why ECMO team members (physicians, bedside nurses, perfusionists, ECMO specialists, and others) prefer one strategy over the other. Why would you extubate a patient on ECMO (“extubation comes first”)? Or why would you prefer to stop ECMO while continuing MV (“discontinuation of ECMO comes first”)? (You can select more than one answer) Responded: 253 Answers in total: 423 Skipped: 0 **Percentage of responders not answers a. Extubation comes first because it decreases the risk for ventilator-associated pneumonia 65 26% ** b. Extubation comes first because it is associated with less muscle (including diaphragm) weakness 72 28% ** c. Extubation comes first because it allows optimal neurological assessment and contact between the patient and the relatives 65 26% ** d. Discontinuation of ECMO comes first because it allows stopping anticoagulation 82 32% ** e. Discontinuation of ECMO comes first because managing a patient on ECMO awake and spontaneously breathing is too difficult and risky 80 32% ** f. Other (please specify) 59 23% ** ARDS, adult (or acute) respiratory distress syndrome; COPD, chronic obstructive respiratory deisease; VV ECMO, venovenous extracorporeal membrane oxygenation. Discussion This EuroELSO survey showed significant heterogeneity in sedation management IMV and VV ECMO weaning practices between participating centers. Only one third of participants considered liberation from IMV before weaning of ECMO, demonstrating the current perception of the practicality and feasibility of awake ECMO. The weaning process from extracorporeal respiratory support begins at the moment of the cannulation and continues as arbitrary milestone have been achieved (recovery of tidal volume [VT], improvement of the natural lung, and resolution of the underlying disease). Improvement in lung function and gas exchange should be monitored, and the weaning process should be focused on keeping the balance between native and membrane lung.11 Although the importance of adequate gas exchange during VV ECMO and the contribution of and synergy between the native and membrane lung are well known, the optimal weaning strategy still remains unclear.9,11 The main adjustments are the ECMO blood flow, sweep gas flow, and sweep gas fraction of delivered oxygen. Weaning from IMV support in VV ECMO is a continuous process which does not necessarily include decannulation from ECMO.9,11,19,20 VV ECMO allows for the reduction in the mechanical power forced to the lung by IMV, thus providing improved lung protection. There is increasing interest to explore the practice of liberation from IMV during ECMO.3–6,21,22 The evidence and data are limited concerning which patients may benefit from IMV weaning before ECMO decannulation as well as supportive protocols for guiding this “combined” weaning processes.8,22 Althoguh our survey shows that in most (58%) centers ECMO decannulation occurs before extubation, some important, yet malleable factors were used for reasoning. Lack of expertise in managing extubated patients on ECMO, institutional protocols and a clinicians’ personal preference of not extubating during ECMO are reasons influenced by enhancing education in ECMO and building evidence of successful extubation during ECMO. Although the stated goal for many centers is to extubate patients while on ECMO, most centers extubate less than half their VV ECMO patients before ECMO decannulation (Q8). Additional reasons were given in free text to the multiple-choice answers (Q13). Responders try to discontinue the most invasive therapy first, because of the multiple potential complications (e.g., infection, bleeding). It was also specified as “most patients were not in a condition to accept ECMO without sedation.” Another reason not favoring extubation during ECMO expressed was: “ECMO is usually shorter if discontinued before weaning the ventilator, which may take several days after weaning ECMO. And the risk for the patient and the burden for the team is lower the sooner the patient is weaned from ECMO.” Extubation on VV ECMO may be postponed due to real or presumed risk for failure expressed by team members (“we consider extubation, but not always feasible”), for example, persistent hypoxemia on ECMO, resource limitations, patient habitus, hemodynamic instability, mucus plugs, etc. Further research is needed to assess the decision-making process of weaning IMV before ECMO, if weaning from ECMO is more beneficial than weaning from IMV and if it is specific to certain patient populations. The main limitation of the study is the inclusion bias caused by the research methodology. The authors may assume that providers are answering the questions based on authors’ perception of the overall strategy used at their institution. The generalizability of the data is also limited because of the fact that patients on VV ECMO is inherently a heterogenous group that may demand approaches based on a number of individual factors. Further, results are presented as descriptive numbers and percentage. No statistical analysis concerning centers’ characteristics, patient categories, etc. could be performed due to the survey database design. On the contrary, the scope of this study was to describe the actual approach used by ECMO providers. Conclusion The variation in the management strategies between centers may be influenced by case mix, experience and volume. Only 32% of providers perceive that their center pursues extubation before weaning ECMO; 31% of participants report their weaning strategy to be similar regardless of reason for respiratory failure, whereas the remaining centers tend to use a varied approach from case to case. Some patients cannot be safely extubated while on VV ECMO while others are clear candidates. Adequacy of participation in pulmonary hygiene (deep breath, cough) is paramount and limited sedation with adequate pain and anxiety management are required. Ambulation is the Elysian ideal. We suggest further research on how clinicians and/or center decide in favor of weaning IMV before ECMO. There is also a need for research to define whether weaning from ECMO before weaning from ventilation is superior to weaning from ventilation before liberation from ECMO. Very likely, different patient populations will benefit from an individualized approach. Acknowledgment Dr. Kiran Shekar acknowledges Research Fellowship support from Metro North Hospital and Health Service.

Topics & Concepts