Research

Preliminary Communication | CARING FOR THE CRITICALLY ILL PATIENT

Effect of Noninvasive Ventilation Delivered by Helmet
vs Face Mask on the Rate of Endotracheal Intubation
in Patients With Acute Respiratory Distress Syndrome
A Randomized Clinical Trial
Bhakti K. Patel, MD; Krysta S. Wolfe, MD; Anne S. Pohlman, MSN; Jesse B. Hall, MD; John P. Kress, MD

Editorial page 2401
IMPORTANCE Noninvasive ventilation (NIV) with a face mask is relatively ineffective at Video at jama.com
preventing endotracheal intubation in patients with acute respiratory distress syndrome
Supplemental content at
(ARDS). Delivery of NIV with a helmet may be a superior strategy for these patients.
jama.com

OBJECTIVE To determine whether NIV delivered by helmet improves intubation rate among CME Quiz at
patients with ARDS. jamanetworkcme.com and
CME Questions page 2463

DESIGN, SETTING, AND PARTICIPANTS Single-center randomized clinical trial of 83 patients
with ARDS requiring NIV delivered by face mask for at least 8 hours while in the medical
intensive care unit at the University of Chicago between October 3, 2012, through September
21, 2015.

INTERVENTIONS Patients were randomly assigned to continue face mask NIV or switch to a
helmet for NIV support for a planned enrollment of 206 patients (103 patients per group).
The helmet is a transparent hood that covers the entire head of the patient and has a rubber
collar neck seal. Early trial termination resulted in 44 patients randomized to the helmet
group and 39 to the face mask group.

MAIN OUTCOMES AND MEASURES The primary outcome was the proportion of patients who
required endotracheal intubation. Secondary outcomes included 28-day invasive
ventilator–free days (ie, days alive without mechanical ventilation), duration of ICU and
hospital length of stay, and hospital and 90-day mortality.

RESULTS Eighty-three patients (45% women; median age, 59 years; median Acute
Physiology and Chronic Health Evaluation [APACHE] II score, 26) were included in the
analysis after the trial was stopped early based on predefined criteria for efficacy. The
intubation rate was 61.5% (n = 24) for the face mask group and 18.2% (n = 8) for the helmet
group (absolute difference, −43.3%; 95% CI, −62.4% to −24.3%; P < .001). The number of
ventilator-free days was significantly higher in the helmet group (28 vs 12.5, P < .001). At 90
days, 15 patients (34.1%) in the helmet group died compared with 22 patients (56.4%) in the
face mask group (absolute difference, −22.3%; 95% CI, −43.3 to −1.4; P = .02). Adverse
events included 3 interface-related skin ulcers for each group (ie, 7.6% in the face mask group
had nose ulcers and 6.8% in the helmet group had neck ulcers).

CONCLUSIONS AND RELEVANCE Among patients with ARDS, treatment with helmet NIV
resulted in a significant reduction of intubation rates. There was also a statistically significant
reduction in 90-day mortality with helmet NIV. Multicenter studies are needed to replicate
these findings. Author Affiliations: University of
Chicago, Department of Medicine,
Section of Pulmonary and Critical
TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT01680783 Care, Chicago, Illinois.
Corresponding Author: John P.
Kress, MD, Section of Pulmonary and
Critical Care, University of Chicago,
5841 S Maryland Ave, MC6026,
JAMA. 2016;315(22):2435-2441. doi:10.1001/jama.2016.6338 Chicago, IL 60637 (jkress@medicine
Published online May 15, 2016. .bsd.uchicago.edu).

(Reprinted) 2435

Copyright 2016 American Medical Association. All rights reserved.

Downloaded From: http://jamanetwork.com/ on 12/24/2016

The helmet could directly comparing face mask to helmet NIV for the preven. We conducted a cause it requires 2 port sites for inspiratory and expiratory flow. oxygen saturation <88%. were of assist-control mode with tidal volumes of 6 mL/kg of ideal pregnant. Inspiratory pressure was increased in increments of 2 to 3 cm H2O to obtain a respiratory rate of less than 25/min and disappearance of accessory muscle activity. The helmet. This could reduce intubation rates lay set to 50% of maximal inspiratory flow (Video). or upper airway obstruction.5 Although benefits of face mask NIV for continue with the face mask (control) or switch to a helmet in- chronic obstructive pulmonary disease (COPD) exacerbations6 terface (intervention). However. respiratory rate >36/min). gen saturation to more than 90% at an inspired oxygen frac- tient outcomes. those findings have not been replicated. Patients who re- sence of airway protective gag reflex. A computer-generated. if possible. The helmet group received NIV via an ICU ventila- About half of patients with hypoxemia. Patient de. no involvement from the research team. Research Preliminary Communication Helmet vs Face Mask and Endotracheal Intubation in Patients With ARDS N oninvasive ventilation (NIV) by face mask can obviate mographics such as race were collected by self-report with fixed the need for endotracheal intubation and improve mor. To avoid carbon dioxide rebreathing. a Glasgow coma scale score lower than 8. the ventilator pres- sign may allow increased titration of positive airway pressures surization time was set to 50 milliseconds and cycling off de- without substantial air leak.19 These included neurologic deterioration. made helped with face mask ventilation.4 Noninvasive ventilation allows pa. cisions to intubate were made by the primary care team with monary arrest. Each assignment was designated in a nocompromised patients with hypoxemic respiratory failure8. The block versial. intolerance of face mask or hel- fined by the Berlin criteria. however. for consent. producing a breathing circuit closed from the outside environ- fore. not be managed with the Philips Respironics V60 ventilator be- tion of endotracheal intubation in ARDS.com/ on 12/24/2016 . The face mask group was managed with a single-limb non- To our knowledge. are not or continuous positive airway pressure mode.12 Often higher levels of of transparent latex-free polyvinyl chloride. 2016 Volume 315. Written informed consent was obtained The decision to intubate all patients was based on prede- from participants or from their authorized surrogate decision termined criteria similar to those used in previous studies of maker. Race data were collected to reflect the diversity of tality in patients with acute respiratory failure. especially those tor (Engström Carestation. support was reduced progressively Methods in accordance to clinical improvement and discontinued if the Consecutive patients admitted to the adult medical ICU at the patient maintained a respiratory rate of less than 30/min and University of Chicago from September 2012 through Septem. The hel- An alternative is to deliver NIV via a helmet interface—a met was connected to the ventilator by conventional respira- transparent hood that covers the entire head of the patient with tory circuitry joining 2 port sites to allow inspiratory and expi- a soft collar neck seal. airway bleeding. ab. For NIV weaning.1 excessive sedation. consecutively numbered. a plastic ring connecting the helmet to a latex-free neck seal.18 and extend the benefits of NIV to more patients with ARDS. there have been no randomized trials invasive ventilator (Philips Respironics V60). the face mask has limitations that may contribute to ment.9 A study by Patients randomized to the intervention switched from Frat et al10 showed increased mortality was associated with face a face mask (Philips Respironics) to a latex-free helmet mask NIV for AHRF compared with high-flow nasal cannula.16 met. thus ance and air leak can impede effective oxygenation. Com.9 neck seal was cut to ensure a tight but comfortable seal. and optimize patient-ventilator synchrony. single-center. sealed. categories. quired endotracheal intubation had initial ventilator settings sure. allocation was blinded. permuted block and cardiogenic pulmonary edema7 are compelling.17 To minimize inspiratory effort proved seal integrity at the neck. GE Healthcare) in pressure support with acute respiratory distress syndrome (ARDS). was secured by pad- positive end-expiratory pressure (PEEP) to improve oxygen. (Sea Long). 2 delirium.11.15 Therefore. They were randomly assigned in a 1:1 ratio to either opted in many ICUs. The patient neck circumference was measured and the reduced efficacy during AHRF. 4 to 8 was used to allocate patients to each group. randomized clinical trial of patients admitted to Both groups had titration of NIV by a standard protocol: PEEP the ICU for ARDS requiring NIV to determine whether helmet was increased in increments of 2 to 3 cm H2O to improve oxy- NIV could reduce the rate of intubation and improve other pa.com Copyright 2016 American Medical Association. elevated intracranial pres. Number 22 (Reprinted) jama. opaque envelope. Initial reports suggested improved survival in immu. ratory flow. a strategy now ad. All rights reserved. rate of more than 100 L/min. tracheostomy. This interface confers several advan. Patients 18 years or older who required face mask NIV NIV. pressure tages over face mask including improved tolerability and less support levels were set to maintain a ventilator inspiratory flow air leak due to the helmet’s lack of contact with the face and im. The institutional review board FIo2 less than 50% and PEEP of less than 5 cm H2O. approved the study. Acute respiratory distress syndrome was de. patients admitted to the medical ICU.10. plications of endotracheal intubation include pneumonia.13 There. After 8 hours of NIV via face mask. face mask intoler. for enrollment. 3 and intensive care unit Intervention (ICU)–acquired weakness.14. persistent for at least 8 hours for the management of ARDS were eligible or worsening respiratory failure (eg. patients were approached tients to remain animated while in the ICU. ded armpit braces attached to 2 hooks on the front and back of ation are needed. at high PEEP. partial pressure of oxygen (PaO2) of more than 75 mm Hg with ber 2015 were screened daily. the helmet’s de. tion (FIO2) of 60% or less. All de- Patients were excluded if they had impending cardiopul. Downloaded From: http://jamanetwork. or copious respiratory secretions. body weight20 and titration of PEEP to achieve oxygen satu- 2436 JAMA June 14. or had refused endotracheal intubation. its use in randomization scheme with varying block sizes ranging from acute hypoxemic respiratory failure (AHRF) remains contro.

26 To evaluate the ef.2 awakening and breathing trials.0 (StataCorp LP) software both efficacy and safety after the enrollment of 83 patients. 9 Upper airway obstruction ment for multiple comparisons. All rights reserved. claustrophobia).23. The area under the curve was calculated for for early stoppage of the trial for efficacy. blind. of whom 83 patients were randomized and ses were planned at one-third and two-thirds of enrollment enrolled (Figure 1). Thirty-nine patients were assigned to jama. criteria 456 Received NIV for <8 h derwent endotracheal intubation based on criteria established 5 Previously enrolled a priori. Number 22 2437 Copyright 2016 American Medical Association. patient intolerance (ie. oxygen saturation. FIO2. Patients who died during the study were assigned scores of 0 for ventilator-free days. DSMB determined that there were no safety concerns and that The effect of the intervention was compared between groups the study had not met other secondary end points that (eg. Helmet vs Face Mask and Endotracheal Intubation in Patients With ARDS Preliminary Communication Research ration of 88% to 95% at lowest possible FIO2 (goal FIO2 <0. the DSMB every measured respiratory rate. in current study 40 Hypercarbic lator-free days (ie. Because of the nature of the 2 intervention groups. From October 2012 through September 2015.6). was predetermined at a P value <. 115 Met ≥1 exclusion criteria 85 Do-not-intubate day mortality. respectively). respiratory failure duration of ICU and hospital length of stay. a time-to-event patients would not be deprived of the benefit. ICU assessed for eligibility tors specific to helmet NIV use and included skin ulceration at the neck seal. however. 2016 Volume 315. nonstudy fect of the intervention on 90-day survival. ICU using the log-rank test.001 for rejection of the null cluding the primary outcome.19 Secondary outcomes were 28-day invasive venti. face mask. the results met criteria tinuous variables. Results Safety Monitoring An independent data and safety monitoring board (DSMB) con. Downloaded From: http://jamanetwork. study already had met the preestablished criteria for early stop- vention. Adverse events were prespecified to include fac. the analysis estimated with the Kaplan-Meier procedure was used. Early stopping for efficacy was used as appropriate to compare categorical variables. PEEP. therefore.24 we calculated that 83 Randomized enrollment of a total of 206 patients would provide 80% power to detect a 20% absolute reduction of the primary outcome. determined that the trial should continue because the helmet and pressure support levels during NIV. and 740 Had acute respiratory failure requiring noninvasive ventilation device complications (ie. were screened.21 and early mobilization22 were performed per ICU stan- 4226 Patients admitted to the medical dard care. page.24 we reasoned that a 30% intuba- tion rate (ie. the DSMB recommended that the study be stopped for timated using this model. helmet deflation). Interim analy. NIV. Figure 1. hospital and 90.27 treated with face mask NIV compared with high-flow nasal can- Additional analyses were performed with Cox-regression nula. sis. 657 Excluded Study Outcomes 501 Did not meet entry The primary outcome was the proportion of patients who un. The DSMB determined that the face mask group could have models that adjusted for Acute Physiology and Chronic Health been exposed to increased risk of mortality and because the Evaluation (APACHE) II and the presence of the helmet inter. Because we have multiple order secondary outcomes. a 20% reduction) would be a clinically significant 39 Included in primary analysis 44 Included in primary analysis improvement. the DSMB evaluated work by each randomized group was estimated using a nonparamet. days alive without mechanical ventilation). All analyses were performed by an intention-to-treat analy. The cumulative incidence of intuba. Hazard ratios (HRs) together with 95% CIs were es.com (Reprinted) JAMA June 14.com/ on 12/24/2016 . we considered them explor. 1 Cardiac arrest atory. was used for statistical analyses. The study protocol and sta- tistical analysis plan are available in the Supplement. with 39 Randomized to receive NIV 44 Randomized to receive NIV via face mask via helmet a 2-sided α level of . Subsequent to this. At the first interim analysis. Because previous work has shown that 39 Received NIV via face mask 44 Received NIV via helmet 50% of patients with ARDS treated with NIV delivered via face as randomized as randomized mask required intubation. Frat et al10 that reported increased mortality among patients ric estimator and compared using the Fine-Gray test. 15 No research staff available 41 Declined to participate Statistical Analysis Assuming an intubation rate of 50% for patients with hypox- emic respiratory failure requiring NIV. Wilcoxon-Mann-Whitney hypothesis to declare that the helmet strategy was superior to 2-sample rank sum test or t tests were used to compare con. length of stay) could have been reached with further enroll- tion (with death without intubation as a competing risk) within ment. Stata 11. and adverse events. In addition. Flow of Participants Through Study Daily interruption of sedation. and we analyzed them without adjust. 740 patients tinuously monitored safety and study conduct. in. noninvasive ventilation. was not available for use outside the trial.10. 3 No gag reflex 2 Pregnant ing was not possible for the outcomes of interest.25 The χ2 test or Fisher exact test (70 and 140 patients. ICU indicates intensive care unit.05.10.

5/min after ran- Extrapulmonary ARDS 6 (15) 3 (7) domization (absolute difference.4. Sixty patients 0. Patients in the helmet group had a me- White. Hispanic 3 (8) 3 (7) group (absolute between-group difference. absolute difference.92 days. Number 22 (Reprinted) jama.02). tachypnea and Steroid use 15 (38) 23 (52) hypoxemia despite protocolized adjustment of NIV settings (83. body mass −45. −8.2% Inspiratory positive airway 10 (10-15) 12 (10-14. 24. 12. 95% CI. 28. In contrast.15. 0. The median interval of NIV prior to random.4) Primary and Secondary Outcomes randomization. pressure. the helmet group (active or in remission <5 y).4 to 3. 95% CI. In contrast. Both groups had was also independently associated with death at 90 days (HR. After adjusting for the APACHE II PaO2:FIO2 144 (90-223) 118 (93-170) score and the intervention.21). fraction of inspired oxygen.com Copyright 2016 American Medical Association.4-71. No patient was P = . −6.2/min. (%) 12 (31) 9 (20) remained significant (HR. median (IQR).47 (95% CI.5. −0. % 95 (91-99) 97 (95-99) for the primary outcome of endotracheal intubation was 0.08.2/min to Pneumonia due to 14 (36) 15 (34) immunosuppressionb 6. to 2. P = . interquartile range. P = . 8.5. 0. 1. cm H2O analysis.2 days for the ization was not different between face mask and helmet (13. h The intubation rate was 61. APACHE.03) in the helmet group.63 days. peripheral oxygen saturation by pulse oximeter. the hospital (10. Reason for Acute Respiratory Failure −4. 95% CI.3/min.3/min.8) Treatments Women 18 (46) 20 (45) Patients in both groups had similar postrandomization dura- Black 22 (56) 28 (64) tions of NIV treatment. median. partial pressure of oxygen. P < .4% to −24.5% in the face mask group and 18. The risk of death at 90 days 2438 JAMA June 14.5% for helmet.7/min at baseline to 24. increased risk of death. PaO2. 95% CI.3%. Median (IQR) Duration of NIV before 13 (8-19.001). 95% CI.0 face mask group.4. The 2 groups had statistically similar oxygen Body mass index.91 days. was the most common reason for intubation in the helmet group a Scores on APACHE II range from 0 to 71. 1. Medications The most common reason for intubation among patients in the Pressor requirement 4 (10) 1 (2) face mask group was respiratory failure—ie.3 (8. The APACHE II score (72%) had a PaO2/FIO2 ratio of less than 200. 3. P = . absolute difference.0 H2O vs 5.7) 10.3. conventional face mask and 44 to helmet NIV. absolute difference. Abbreviations: ARDS. −14. −43. Respiratory and Hemodynamic Parameter. 95% CI. y 60.76.3%.1.8/min. % 60 (50-80) 60 (40-90) (95% CI. 95% CI.5 to −9.001). IQR. −7. Titration of PEEP to higher levels per proto- Solid cancer 10 (26) 5 (11) col in the face mask group was limited because of patient in- Hematologic cancer 6 (15) 7 (16) tolerance and excess air leak.2% for face mask.02). P = . The helmet group spent 4. FIO2. 0.5% for helmet vs 4. and about one- No.8. b Immunosuppression was defined as hematologic malignancy or solid tumor In the exploratory secondary analyses.7 days in the ICU vs 7.3-13.001). (%) of Patients Receiving third in each group had an immunocompromised pneumo- Noninvasive Ventilation Face Mask Helmet nia (Table 1).3 hours.9/min to 3. with higher scores indicating (62. −2. 95% CI.47.1) 58 (49.3% for face mask vs 37.8-92.01 to 1. Acute Physiology and Chronic Health Evaluation. 1. absolute difference.6- Asian 1 (3) 2 (5) 2.8. a high severity of illness as indicated by APACHE II scores.27the unadjusted subhazard ratio for the helmet group SpO2. 0. Hospital and 90-day mortality were significantly lower in Characteristics at Inclusion the helmet group than in the face mask group (Table 2). 95% CI. P < . −82.001. unadjusted HR for death at 90 days was 0. In a competing risk Expiratory positive airway 5 (5-8) 5 (5-8) pressure.24 to line characteristics of patients in both groups.1/min. P = .24.3.9.8 days for the face mask group (absolute difference.2 Medical History to −0. Research Preliminary Communication Helmet vs Face Mask and Endotracheal Intubation in Patients With ARDS About half of the patients in each group were immunocom- Table 1.1 days for the helmet group vs 15.5) in the helmet group (absolute difference.001). long-term (>30 d) steroid use of more than 20 mg/d. cm H2O −62. the subhazard score for the helmet Shock.006). P = . median (IQR) 26 (23-30) 25 (20-28) the face mask group (absolute difference. P = . 0.001).8-67. to after ran- Stem cell transplant 1 (3) 5 (11) domization. 58.22 FIO2.9 (56. 29. All rights reserved. Characteristics of Patients at Baseline promised by virtue of cancer or transplant. There was no significant change in respiratory rate in patients receiving NIV via face mask at Solid organ transplant 3 (8) 5 (11) the time of randomization (baseline.54. or use of any immunosuppressive drug for more had more ventilator-free days than the face mask group (28 vs than 30 days. P = . neurologic fail- index.16). median (IQR) 28 (23-35) 27 (24-36) saturations. P < . P = <. acute respiratory distress syndrome. ure (ie. 2016 Volume 315.50. BMI. altered mental status.04) but did not spend statistically significant less time in lost to follow-up. P < . The helmet group had an FIO2 of 50% vs 60% in APACHE IIa. P = . solid organ transplant. the transition from Pneumonia 14 (36) 23 (52) face mask to helmet resulted in a significant reduction in Aspiration 5 (13) 3 (7) tachypnea from 27.11-0.01). non-Hispanic 13 (33) 11 (25) dian sustained PEEP of 8. Characteristic (n = 39) (n = 44) Age.07 to 0.1 cm H2O in the face mask White. 0.7 (95% CI.com/ on 12/24/2016 . 95% CI. absolute difference.1/min. No. Downloaded From: http://jamanetwork. Table 2.11-0. The There was no statistically significant difference between base. loss of airway protective reflex) SpO2. −2.47 vs 10.65). 95% CI. 13.

7) 10.4) <.3 (−82. No.001 Secondary outcomes. d No. There were respiratory failure because endotracheal intubation is asso- 2 instances when the helmet was deflated. 0.com/ on 12/24/2016 .7) −2. corrected and did not result in endotracheal intubation.5-8.8%) in the hel.29 Such complications was no statistical difference in the rate of mask-related skin can have long-standing consequences.6) 3 (6.5 (0.3) 3 (37. intensive care unit.5) 0 (0) −12. Primary and Secondary Outcomes and Adverse Events Absolute Face Mask Helmet Difference (n = 39) (n = 44) (95% CI) P Value Primary outcome. Mask deflation 0 (0) 2 (4.9) −2.3) <.50 Face mask NIV 0.5 to −9.5) a 90-d Mortality includes hospital Skin ulceration 3 (7. NIV delivered hypothesized that the helmet’s neck seal would allow for by helmet significantly reduced the intubation rate among delivery of higher airway pressures without substantial air jama.16 Mortality. this ensured that only those with high illness acuity and a substantial chance of requiring endotracheal intubation were enrolled.4 to 3.2 (7.7 to 0. (%) Hospital 19 (48.7 (2.8 to 92.8 (3.01 Circulatory failure 3 (12. Adverse Events Avoiding intubation is critical for patients with acute Overall. Helmet vs Face Mask and Endotracheal Intubation in Patients With ARDS Preliminary Communication Research Table 2.07 to 0.54) .7) 12 (27. The helmet also was associated with improved 0. 95% CI.1 (6. particularly among ulceration between groups with 3 patients (7.5) 8 (18.9-13.23 to by face mask.5-15. Probability of Survival From Randomization to Day 90 1.8-19.001 Reason for intubation Respiratory failure 20 (83.1) −22.3 (−62. We In this single-center.5 (−25. at risk Face mask 39 20 18 18 18 18 17 Helmet 44 33 31 29 29 29 29 NIV indicates noninvasive ventilation. 2016 Volume 315.75 Probability of Survival Helmet NIV 0.4 (−41. Figure 2.25 Log-rank P =.47 to 2. patients needing NIV for only a short time. ventilator-free days and significantly reduced ICU length of stay as well as 90-day mortality.92 (−8. The sig- nificant reduction in the intubation rate may be explained Discussion in part by the effective delivery of higher levels of PEEP.047.02 0 0 15 30 45 60 75 90 Time From Randomization. (%) Endotracheal intubation 24 (61.9 to −1.1) .02 Adverse events Abbreviations: ICU. 0. randomized clinical trial. patients with ARDS compared with the patients receiving NIV justment for APACHE II score ratio (HR. the incidence of adverse events was low.49-28) 28 (13. P = .7-28) 8.28 respiratory.2) −43. There and musculoskeletal complications. remained significantly lower in the helmet NIV group after ad. All rights reserved.04 Hospital length of stay 15. d Ventilator-free days 12.com (Reprinted) JAMA June 14.99. Downloaded From: http://jamanetwork. which was quickly ciated with numerous infectious1 neurologic.8) 4. IQR.3 to −1. Number 22 2439 Copyright 2016 American Medical Association.4 to −24.30 The 8-hour period of face mask NIV mask group with a nose ulcer and 3 patients (6.3) −21. median (IQR).8) .001 ICU length of stay 7.7) .8) mortality. No.55 Neurologic failure 1 (4. Figure 2).6%) in the face patients with ARDS.2) 5 (62.5) −45.4) .4) 15 (34.63) . interquartile range.3 (24.51.5) 58.76 (−6.00 0.0) .04 90 da 22 (56.3 (−43.4 (13. was chosen a priori as a study entry criterion to avoid met group with a neck ulcer.

typically between 100 to 200 L/min.3 (92. noninvasive (n = 39) (n = 44) P Value ventilation. although there were no differences in over.0-60. as measured by av.0) 8 (5. there is likely to PEEP effects. These higher sustained PEEP lev. SpO2. high ventilator fresh gas flow with the helmet be a learning curve as clinicians gain familiarity. breaths/min 0. the nature of this trial intervention made blind- tus changes and loss of the airway protective reflex.9. Multicenter studies are needed to replicate these findings.17 Thus. Research Preliminary Communication Helmet vs Face Mask and Endotracheal Intubation in Patients With ARDS Table 3. These findings also affirm between 5. should have pressure support levels set to ensure high inspi- sponded with a significant reduction in the respiratory rate ratory flow levels (ie. severity of illness than patients in our trial. the absolute difference was 3.1-34. and future studies replicating this trial may report lower all intubation rates.13 group. P <.17 as well as periodic arterial blood sampling els appear to have maintained acceptable gas exchange.30 In contrast.9 compared with a SOFA score of 7 in the current study. teria for endotracheal intubation to decrease bias.6-10. nurses.7) 96.34 thereby reducing the need for intubation. This corre.9/min to 3.9%. These scores predict a hospital mortality of tubation by enhanced PEEP effect. Level of Respiratory Support and Physiologic Parameters During Noninvasive Ventilation Noninvasive Ventilation.0) 19. for the helmet group. Careful train- interface was noted.com Copyright 2016 American Medical Association. leak.6) 24. Median (IQR) Abbreviations: FIO2. Number 22 (Reprinted) jama. % 95.21). 2440 JAMA June 14. the absolute difference was Respiratory rate.33 Lemiale et al9 noted no difference the far-reaching benefits of spontaneous yet highly sup- in intubation or 28-day mortality in immunocompromised pa. Conclusions The helmet interface is a relatively novel approach to NIV For patients with ARDS. in the helmet. tion to long-term outcomes. Third. PEEP. These find- Despite being immunocompromised.35 Also recruitment maneuvers cannot be ap. we followed predetermined cri- The observed intubation and mortality rates among pa. with predicted mortality rates in the based on predetermined criteria.68 a Median area under the curve of PEEP.0-8. although this study was stopped early for efficacy high APACHE II scores. early stoppage of face mask NIV compared with patients randomized to high. during helmet use. greater than 100 L/min—this was al- and similar oxygen saturation levels on a lower FIO2 than ways easily achievable with modest pressure support set- achieved with face mask.0-60.4)b 27. but instead usually due to mental sta. Downloaded From: http://jamanetwork. to respiratory failure.0) <. treatment with helmet NIV was asso- and this study has several cautions and limitations.0-68.4 (7.10. Pressure support.4-45. as tients in the face mask group were higher than some recently a single-center trial. ported ventilation in an awake. the significance of the 50% range.8 (8. h 26.6) . gest biologic plausibility for the prevention of endotracheal in- tween 24 and 27. All rights reserved. fraction of Face Mask Helmet inspired oxygen.0) . % 60 (50.36 The study findings sug- domized to the helmet group had substantially higher levels gest that patients whose ARDS was managed with helmet NIV of PEEP. First. ing of all physicians and staff will be needed.001).5) 8 (5.1/min.2/min After randomization 29. trials tends to exaggerate the magnitude of the effect size flow nasal cannula. In addition to the Second. In the exploratory secondary analyses.02 randomization respiratory rates within groups: for the face mask SpO2. just as was the High fresh gas flow rates reduce the risk of CO2 rebreathing case 20 years ago when face mask NIV was first introduced.2 (10. which were sustained throughout NIV.3/ Baseline 28. animated patient over inva- tients randomized to receive mask NIV or oxygen therapy alone.4) . cm H2O 11. the PEEP and fresh gas flow effects of Physicians. −4.4-30. The patients in this trial had much lower efficacy of helmet NIV.5) (95% CI.1-37. our results may not have external valid- reported studies of AHRF. 2016 Volume 315. similar to the face mask NIV group.1 (5.6) 50 (40. tings. Fifth. Antonelli et al24 previously reported that pa- tients with ARDS treated with face mask NIV had a 51% intu- bation rate and a 64% hospital mortality. positive end- Respiratory support with NIVa expiratory pressure. sive mechanical ventilation via endotracheal tube. like any new tool or technology.0-14.3-96. Duration of NIV.com/ on 12/24/2016 . particularly with atten- groups. There was also a statisti- may lead to C O 2 rebreathing 17. peripheral oxygen saturation by pulse oximeter.3 (22. effect size of the primary outcome suggests that the prob- creased mortality among patients with AHRF randomized to ability of type I error is very low.5-34.0-10.7 (21.8-98. However. ing impossible.32 A study by Frat et al10 recently reported in. The physiologic effects observed with helmet NIV sug- erage Simplified Acute Physiology Scores (SAPS II) of be.2/min to 6. cm H2O 5. Fourth. 0. and respiratory therapists involved in this helmet NIV appear to have improved oxygenation and work study quickly became familiar and comfortable with helmet of breathing so that failures of helmet NIV were rarely due NIV during the course of the trial.5 (20.8% and 7. Accordingly.31 However.2 (94.0) . patients ran.6)b min.001 b Comparison of baseline and after FIO2.34 and patient-ventilator cally significant reduction in 90-day mortality with helmet NIV. NIV.1 (22.8/min (95% CI. the median admission ings warrant further investigation of helmet NIV for patients Sequential Organ Failure Assessment (SOFA) score was 5 in both with ARDS and other types of AHRF. plied with noninvasive ventilation. P = . the ciated with a significant reduction of intubation rates com- large internal volume of the helmet and its high compliance pared with delivery by face mask. Table 3).006 respiratory support. our patients had very ity. dyssynchrony.

Bufi M. Bellani G.339(7):429-435. du patient d’Onco-Hématologie (GRRR-OH).270(24):2957-2963. N Engl J Med. ventilator. Intensive Pohlman. syndrome. Campo FR. Pohlman MC. technical. Cochrane protocol for mechanically ventilated patients in Database Syst Rev. Antonelli M. Hill NS. or material support: Patel. 370(17):1626-1635. 2011. Hall. Ventilation with lower tidal volumes as 35. Kocherginsky M. Dignam JJ. N Engl J Med. Sea Long. Campbell MJ. 1990. and Intensive Care Units Network (PICUN). Noninvasive interfaces: should we go to ARDS. Critical revision of the manuscript for important 10. Acute respiratory of adults with hospital-acquired. Am J Respir Crit Care therapy. Functional 1. Early physical and occupational therapy in Author Contributions: Drs Patel and Kress had full access to all of the data in the study and take 9. Hermans G.37(6):2111-2113. Vargas F. Van Mechelen H. LUNG SAFE associated. Antonelli M. Guidelines for the management ARDS Definition Task Force. ARDS Network. Brun-Buisson C. 2009. Le Gall JR. or interpretation of data: Patel. Continuous positive airway pressure delivered with Daily interruption of sedative infusions in critically ill care units in 50 countries. ESICM Trials Group. in patients with acute respiratory failure. Kress. Piedmont cognitive impairment after critical illness. et al.363(12):1176-1180. (14):1753-1762. positive pressure ventilation (CPAP or bilevel NPPV) safety of a paired sedation and ventilator weaning 2013.291 2009. Conflict of Interest Disclosures: All authors have Córdoba-Izquierdo A.333(13):817-822. 34.30(1):147-150. Study concept and design: Patel. et al. doi:10. Navalesi P. Acute outcomes and 1-year mortality of intensive Additional Contributions: We thank respiratory care unit-acquired weakness: a cohort study and 15. Hall JB. Study supervision: Hall. Pantanetti S. Ely EW. fever. analysis. et al. JAMA. et al. Draper EA. Kacmarek RM. Crit Care Med. Atallah AN. A new Simplified Acute Physiology Score (SAPS II) 4. Study Group. 2014. Catani F. JAMA. 2000. 2016. Factors models. Frat JP. Mokart D. 2007. critically ill patients: responsibility for the integrity of the data and the Groupe de Recherche en Réanimation Respiratoire a randomised controlled trial. Kress JP. Long-term 14.307(23):2526-2533. through nasal cannula in acute hypoxemic ventilation in patients with acute hypoxemic Administrative. Crit Care Med. Kress.2016. Canadian Critical Care Trials Group. Resche-Rigon M. Thille AW. Matthews JN. Noninvasive Acquisition. et al. Zhang Q. Antonelli M. intensive care (Awakening and Breathing jama. 11. Helmet with specific settings versus predictor of mortality in mechanically ventilated classification system. 20. analysis. Wolfe. Bigatello LM. Antonelli M.364(14):1293-1304. BRAIN-ICU Study Investigators. et al. et al. N Engl J Med. propensity-matched analysis. The use helmet devices were obtained commercially from and interpretation of competing risks regression 13. Lemeshow S. Noninvasive Controlled trial): a randomised controlled trial. delivered by helmet in hematological malignancy patients with hypoxemic acute respiratory failure. Thompson BT. and Noninvasive continuous positive airway pressure nursing staff for their support of this project. Wagner DP. et al. Tansey CM. 28. 36. a randomized trial. 2004. JAMA. in patients undergoing solid organ transplantation: 22. Contou D. 31. 818-829. 1985. Fellow salary and practice of noninvasive ventilation as a first-line medical research.372(23): respiratory failure: a multi-center study. 2185-2196. Pohlman AS.13(10): facemask for noninvasive ventilation. JAMA. Slutsky AS. Bernard GR. Esquinas A. mortality among immunocompromised patients ventilation in immunosuppressed patients with Wolfe. 2004.49(3):270-275. Clin Cancer Res. Fuchs BD. trial. 2009. 2005. Shintani A. Moro ML. or Med. Ladeira MT. Delirium as a Brochard L. Mancebo J. Thille AW. Lancet. distress syndrome: the Berlin Definition. et al.314(16):1711-1719. et al. and healthcare-associated pneumonia. 30. Altman DG.com/ on 12/24/2016 . Number 22 2441 Copyright 2016 American Medical Association. N Engl J collection. All rights reserved. Kress. Brochard L.300(6719):230-235. Kress. APACHE II: a severity of disease 3. with acute respiratory failure: a randomized clinical pulmonary infiltrates. Rocco M. Society of America. Hall JB. Published Online: May 15. N Engl J Med.5:CD005351.com (Reprinted) JAMA June 14. 2000. patients in the intensive care unit. Investigators. Hess D.344(7):481-487. Kress JP. preparation. Fragnoli C. Jackson JC. Management of 7. 2015. et al. JAMA. et al. and mortality for patients with 17. mechanically ventilated. Lemiale V. 2002. Predictors intellectual content: All authors. trial. Ranieri VM. 32. Boissier F. affecting oxygen delivery with bi-level positive Role of the Funder/Sponsor: Edelman family had airway pressure. 2015. Crit Care Med. no role in the design and conduct of the study. Crit Care. 2012. Taccone P. JAMA. and acute respiratory Drafting of the manuscript: Patel. ventilation for treatment of acute respiratory failure 2008. Care Med. REVA Network. et al. Wolfe.283(2):235-241. Squadrone V. of noninvasive ventilation vs oxygen therapy on 23. Kress. et al.190(4):410-420. Conti G.37(6):1921-1928. the manuscript for publication. 25. using funds from an unrestricted internal grant A multiple-center survey on the use in clinical Royston P. Conti G. Laffey JG. Am J Respir Crit Care Med. 1998. Analysis of serial measurements in from the Daniel J.32(10):2090-2096.315(8):788- a “helmet”: effects on carbon dioxide rebreathing. 2004. N Engl Crit Care Med. Saulnier F. Villar J.18(8):2301-2308. lung injury and the acute respiratory distress Care Med. et al. High-flow oxygen of failure of noninvasive positive pressure Statistical analysis: Patel. N Engl J 5. none were reported.30(8):1772-1777. JAMA. helmets? Crit Care Med.342(18):1301-1308. Respir Care. acute respiratory distress syndrome.369(14):1306-1316. 2013. patterns of care. Conti G. Vital FM. patients undergoing mechanical ventilation. Hall. Epidemiology. Cerutti E. Hilbert G. REFERENCES Intensive Care Med. Clerck B. Kacmarek RM. Knaus WA. Efficacy and refractory hypoxemia in ARDS.27(11):1718-1728. 2013. positive airway pressure for treatment of approval of the manuscript. Kress JP. and decision to submit postoperative hypoxemia: a randomized controlled 29. Pandharipande PP. Truman B. Helmet vs Face Mask and Endotracheal Intubation in Patients With ARDS Preliminary Communication Research ARTICLE INFORMATION 8. Conti G. 800. Kress. Matté A. J Med. 2016. Rubenfeld GD. The Acute Respiratory Distress Syndrome 6.171(4):388-416.6338. Girard TD. N Engl J Med.35(1):18-25. review.1001/jama.342(20):1471-1477. 26. American Thoracic Society. et al. 27. 1995. Hess DR.” Crit chronic obstructive pulmonary disease. FLORALI 24. Helmet to deliver Noninvasive ventilation for acute exacerbations of compared with traditional tidal volumes for acute noninvasive ventilation: “handle with care. Vargas F. N Engl J Med.79(10):1173-1179. acute respiratory distress syndrome in intensive 2. Gruson D. Crit Care Med. ICU-acquired weakness and A comparison of noninvasive positive-pressure recovery from critical illness. BMJ. Infectious Diseases disability 5 years after acute respiratory distress 16. 2000. Schweickert WD. 33. Schoenfeld DA. Edelman family. 2001. All syndrome. Med. Thille A. failure. management. Non-invasive 21. Continuous interpretation of the data. Pohlman. et al. Minerva Anestesiol. 2014. Coha M. 2010. Wysocki M. 2004. et al. Zimmerman JE.373 accuracy of the data analysis. Girard TD. Funding/Support: The helmets were purchased 12. Herridge MS. 2013. Statistical evaluation of ventilator-free days as an completed and submitted the ICMJE Form for Non-invasive ventilation for acute hypoxemic efficacy measure in clinical trials of treatments for Disclosure of Potential Conflicts of Interest and respiratory failure: intubation rate and risk factors. Lyazidi A. Neuromuscular blocking agents in Med.17(6):R269. Network. Pham T. Nava S. 2016 Volume 315. syndrome. Mercat A. respiratory failure. physical and occupational therapists. 19. for cardiogenic pulmonary oedema.293(5):589-595. 2005. Caironi P. N Engl J Med. Effect (9678):1874-1882. O’Connor MF. Lancet.32(10):2162-2163. Principi T. 18.371(9607):126-134. 2004. 1993. 2001. based on a European/North American multicenter ventilation and conventional mechanical ventilation study. Schwartz AR. support for Drs Patel and Wolfe was provided by intervention for acute respiratory distress grant T32 HL007605 from the NIH/NHLBI. 2012. Downloaded From: http://jamanetwork. Pohlman AS.