FAQs
Can’t find what you’re looking for? Submit a question to the WINDSURFER team!
WINDSURFER is designed to evaluate which non-invasive respiratory support (NIRS) modality emergency clinicians should reach for when treating Emergency Department patients with acute hypoxemic respiratory failure (AHRF). The trial focuses on the earliest phase of care when phenotype is often uncertain.
Key points about WINDSURFER:
- Enroll all patients with acute hypoxemia and tachypnea. The trial does not target any particular etiology.
- Enroll as soon as it is known that NIRS is needed. For patients that meet the simple inclusion and exclusion criteria, we will not wait for further information or clinical change.
- There is equipoise regarding HFNO and NIPPV effectiveness in all causes of hypoxemia, whether COPD, heart failure, obesity-hypoventilation, etc. Do not limit enrollment based on past guidelines, preferences, or biases.
- If urgent intubation is needed, then do not enroll. WINDSURFER is not a pre-oxygenation study.
Overall, the general rule of thumb is to:
Enroll when the patient needs NIRS, do NOT wait until you know the etiology of their respiratory failure.
Q1. Can a patient be enrolled after they have been started on NIRS for clinical care?
A1. Yes, but this should be the exception and should happen rarely. The expectation is that each site will establish a workflow to identify and randomize patients early so the index NIRS is the randomized modality.
There are a few clinical scenarios when NIRS may need to be started before enrollment:
1. Patient arrives on prehospital NIRS,
2. Patient was preoxygenated for intubation but improved and can be randomized.
3. The NIRS to which a patient is randomized is not immediately available, so another NIRS modality is used as a bridge.
Site enrollments are reviewed on a rolling basis, and any site with >50% of their enrollments randomized after temporizing NIRS will be asked to come to the SIREN operations committee or WINDSURFER operations committee to review their enrollments and develop an action plan to improve their process for identifying and randomizing patients pre-NIRS.
Q2. Why does the protocol say patients should be enrolled within 2 hours of need for NIRS but the expectation is enrollment within 1 hour?
A2. The expectation is that patients are enrolled within 1 hour of identification of the need for NIRS so that the first NIRS modality used is the randomized modality. The protocol allows for up to 2 hours for enrollment to account for unexpected clinical or study delays, such as an LAR that is physically present and able to have an informed refusal conversation, etc. If >50% of a site’s enrollments are after 1 hour, that site will come to the SIREN operations committee or WINDSURFER operations committee to review their enrollments and develop an action plan.
Q3. Should I enroll patients with acute decompensated heart failure patients and extreme hypertension that I think might improve quickly with NIPPV (i.e. “SCAPE”)?
A3. This sub-phenotype of patients with acute decompensated heart failure often improves quickly with blood pressure control, not necessarily the positive pressure, and would definitely not be treated solely with positive pressure. Ultimately, these patients are essentially acute decompensated heart failure. WINDSURFER only excludes patients with advanced chronic heart failure, not acute decompensation (i.e., those at the extremes of baseline severity (heart transplant), not acute severity). The primary literature supports HFNO in heart failure.
Q4. What about patients with obesity hypoventilation syndrome (OHS) who are hypoxemic because of hypercapnia from non-compliance with their home NIPPV?
A4. Patients on home NIPPV for OHS, or any cause other than sleep apnea, would be excluded from WINDSURFER.
Q5. What about patients with a pulmonary embolism?
A5. The new PE guidelines recommend HFNO over standard oxygen, but the supporting text states that either HFNO or NIV are ok and supported.
Additionally, a large enough PE to cause RV failure would be excluded by the shock criteria. For everyone else, NIPPV would unlikely tip them over the edge
Q6. What about patients with an isolated asthma attack, or COPD with severe hypercapnia (PCO2 >80) and acidemia (pH <7.2)?
A6. A pH <7.2 approaches the threshold for intubation in all clinical guidelines. Where NIPPV may be recommended, the recommendations are largely based on comparisons between NIPPV and conventional oxygen, not a comparison to HFNO. Data suggests that HFNO can also help clear CO2 in these COPD patients. On the other hand, in patients with an asthma attack of this severity, a PaCO2 >80 or low pH are very concerning for impending respiratory arrest and these patients are usually intubated.
Q7. What about patients with heart failure and large pleural effusions?
A7. Patients with enough volume overload to cause pleural effusions needs diuresis. NIPPV doesn’t treat the pleural effusions, it reduces the work of breathing and improves hypoxemia while patients are diuresed. HFNO can do those same things.
Q8. What about patients with chronic restrictive lung disease such as pulmonary fibrosis?
A8. There is no evidence to suggest that either HFNO or NIPPV are superior for managing patients who have baseline restrictive lung disease and new hypoxemia.
Q9. What about patients with a very high body mass index?
A9. In this population, NIRS has only been evaluated for post-extubation management, not acute management of hypoxemia. There is no evidence to suggest a lack of equipoise in this population. Assuming patients don’t meet the criteria for urgent intubation, they should be included in WINDSURFER.
Q10. What about patients with a pneumothorax?
A.10 There is no evidence to suggest either NIPPV or HFNO is better in the presence of a pneumothorax.
In the WINDSURFER protocol, use of the unassigned NIRS modality is allowed for up to two hours. Any use of the unassigned NIRS modality for >2 hours is considered a cross-over and will be a protocol violation. Quantitative data show equipoise in baseline NIPPV and HFNO use. WINDSURFER sites were selected carefully based on assurance of equipoise and protocol compliance. However, site investigator teams will likely have significant pre-trial and ongoing work to break down the biases and heuristics among all clinical staff that may lead to crossovers and protocol deviations.
Q11. How can we best ensure compliance with the protocol for the intervention?
A11. We will provide a “cheat sheet” with guidelines for titration, weaning, and liberation from NIRS and indications for intubation that can accompany the patient during their 24 hours of protocolized care. Adherence to the protocol during the first 24 hours will likely require significant research team involvement, and/or physical presence to ensure compliance as patients move between clinical teams and across care areas. Successful research teams will have co-investigators from the pertinent clinical care teams champion the protocol for enrolled participants. There will be some crossover-induced protocol deviations due to patient-related factors, and some due to clinical team factors. The protocol deviations due to clinical team factors will require an action plan to prevent future deviations.
A suggested scripted response to clinical teams that want to change the assigned intervention and deviate from the protocol (cross over) is: “In WINDSURFER, HFNO and NIPPV are both appropriate as the initial therapy for AHRF. Unless the patient meets criteria for intubation or is clearly failing despite optimized settings, switching to the non-randomized modality is a protocol deviation. Let’s optimize the assigned modality first.”
Q12. What if a patient is not tolerating one device (interface) but does not need intubation and gets switched to the other?
A12. Interface intolerance, defined as the participant repeatedly removing the mask or cannula, is an indication for intubation, as listed in the Manual of Procedures. However, support can be titrated to improve tolerance as outlined in the titration guidelines, or a patient can be crossed over for < 2 hours while a different interface for the randomized modality (e.g., different mask) is obtained. After 2 hours, it is a protocol deviation.
Q13. What about a patient crossed over from the randomized modality to the other for < 2 hours and refuses to go back on the randomized modality?
A13. This is a protocol deviation if it is longer than 2 hours.
Q14. What if a patient refuses to be intubated so is crossed over in an attempt to “rescue”?
A14. This is a protocol deviation if it is longer than 2 hours.
Q15. What if a patient is getting worse so crossed over to the other for preoxygenation and intubation?
A15. This is not a protocol deviation unless >2hrs.
Q16. What if the admitting team wants to switch to the non-randomized modality?
A16. This is a protocol deviation that needs an action plan.
Q17. What if the admitting team wants HFNO so the patient can go to the floor instead of the ICU?
A17. This is a protocol deviation that needs an action plan. Sites that have a differential pathway for the NIRS modalities will need to carefully evaluate the monitoring criteria for participants (e.g., vital sign frequency, etc.) as part of the readiness checklist to determine if the trial protocol can be carried out wherever patients are admitted (e.g., floor, stepdown), or if they enrolled participants should be admitted to a higher level treatment area (e.g., ICU) for the duration of the intervention.
Q18. What about participants who are randomized to HFNO but use nocturnal CPAP for sleep apnea?
A18. Home CPAP settings would not be expected to support patients with hypoxemia. Those participants should stay on HFNO at night while they have respiratory failure. Once they are liberated from HFNO, they can resume their home CPAP settings at the discretion of the clinical teams.
Q19. What if the patient has qualified for home NIPPV but hasn’t started on it yet?
A.19. If the team knows the patient is on, or supposed to be on, home NIPPV for anything other than sleep apnea, they should be excluded.
Q20. If the patient is intubated, then extubated within the first 24 hours, do they have to go back onto their randomized NIRS?
A20. No, the protocol ends at intubation and post extubation care is based on clinical care.