Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

Noninvasive Positive-Pressure Ventilation for Respiratory Failure after Extubation.

Esteban A, Frutos-Vivar F, Ferguson ND, et al.

New Engl J Med 2004; 350:2452-2460. [abstract]

Reviewed by Thomas Cholis MD, Children's National Medical Center, Washington, DC

Review posted January 10, 2005

---

I. What is being studied?:

The study objective:

Whether noninvasive ventilation could reduce mortality in patients who developed respiratory failure within 48 hours of extubation.

The study design:

This study was a randomized, multicenter prospective study involving 37 centers and 8 countries.

The patients included:

Patients had to be older than 18 years of age, mechanically ventilated for at least 48 hours and successfully electively extubated after meeting extubation criteria. These patients (n=980) were then observed for the next 48 hours for the onset of respiratory failure (defined as meeting ≥ 2 criteria of a) respiratory acidosis, b) clinical signs of increased respiratory effort or fatigue, c) tachypnea for 2 hours, and/or d) hypoxemia. 244 patients met the criteria for acute respiratory failure.

The patients excluded:

Patients who had tracheostomies and those for whom they could not obtain informed consent. Patients immediately needing reintubation after extubation (n=23) were also excluded. A total of 221 patients were randomized.

The interventions compared:

The patients who met study entry criteria were randomized to standard medical therapy (n=107) or noninvasive ventilation (n=114). Standard medical therapy patients received supplemental oxygen, respiratory physiotherapy, and/or bronchodilators. These patients could be crossed over to noninvasive pressure ventilation, which was counted as failure of standard medical therapy (n=28). The noninvasive group had full facial mask ventilation with pressure support (titrated to achieve tidal volumes of > 5 ml/kg), positive end-expiratory pressure, and oxygen to keep oxygen saturations over 90 percent. Noninvasive ventilation was kept on for 4 hours and patients were allowed 15 - 20 minute periods off noninvasive ventilation.

Patients were reintubated if after one hour of one therapy they met the following criteria:

1. Lack of improvement in pH or pCO2
2. Mental status changes
3. Decreased SpO2 to below 85% despite high inspired FiO2
4. Lack of improvement in respiratory fatigue
5. Hypotension unresponsive to volume challenge or vasopressors
6. Copious secretions that could not be adequately cleared or resulted in criteria a, b, or c

Attending physicians made the final decision to reintubate.

The outcomes evaluated:

This study's primary outcome was mortality with the secondary outcome being rate of reintubation and length of ICU stay.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Patients were adequately randomized through a random-number table with opaque envelopes. Randomization used variable block sizes and was stratified for study center and presence or absence of chronic obstructive pulmonary disease.

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Was followup complete?

Yes, all patients who met acute respiratory failure within 48 hours were accounted for in the study.

Yes, followup appeared to be complete, as all patients were followed until discharge from the ICU. No note however was made about patients who may have been re-admitted to the ICU after transfer to the ward.

Were patients analyzed in the groups to which they were randomized?

Yes, patients that were randomized to noninvasive ventilation were analyzed as nonventilated patients. However, due to the cross-over nature of the study standard therapy patients were kept all in one group and then were further analyzed in the "rescue" noninvasive pressure group.

Secondary questions:

3. Were patients, health workers, and study personnel "blind" to treatment?

No.

4. Were the groups similar at the start of the trial?

At the start of the trial, patients had similar severity of illness (SAPS II), age, gender, and reason for respiratory failure. Patients had similar ABGs, physiological evaluations, duration of mechanical ventilation, and assessment of respiratory status at time of extubation among groups.

5. Aside from the experimental intervention, were the groups treated equally?

Few details are provided about other aspects of patent care, such as fluid management, antibiotics, cardiovascular support. Given that this study could not be blinded, the investigators should have provided such details to assure us that no systematic differences in other aspects of care existed that might have influenced the outcome.

The addition of the crossover to noninvasive positive pressure ventilation from the standard medical therapy counted as respiratory failure in the standard medical therapy group. Since the study was not blinded, a certain bias would be difficult to assess. Not surprisingly patients who were on the noninvasive positive pressure ventilation arm had longer period of time between diagnosis of respiratory failure and intubation.

III. What were the results?

1. How large was the treatment effect?

Mortality was higher in the noninvasive ventilation group 25% (28/114; 95% CI 17-34%) died as compared to 14% (15/107)(95% CI 8-23%). These included patients who were assigned to standard medical therapy but received "rescue" noninvasive positive pressure ventilation (3/28 of these patients died). The absolute risk reduction of death for the control group was 11% (95% CI 0.6%-21%) and relative risk of death for the noninvasive groups was 1.78 (95% CI (1.03-3.2). The number needed to harm (NNH) using this therapy was 9 (95% CI 5-163).

On further evaluation the differences appeared to be in those patients who needed reintubation. 21/55 (38%; 95% CI, 26-52%) died in the noninvasive ventilation group as compared to 11/51 (22%, 95% CI 12-36%; p=0.06) in the standard medical therapy group.

At interim analysis, a significant increase in mortality was observed among patients who received noninvasive ventilation (p < 0.05), and the study was stopped according to predetermined criteria.

There was no difference in rate of reintubation between the standard medical therapy group and the noninvasive ventilation group (48%; 95% CI 38-57%) compared to the standard medical group (48%; 95% CI 39-58%). There was a time difference in the interval between reintubation and the diagnosis of respiratory failure in the noninvasive group versus standard medical group (12 hours versus 2 hours; p=0.02).

2. How precise was the estimate of the treatment effect?

The confidence interval does not cross 1 in the relative risk nor 0 in the absolute risk, making the results statistically significant. However, the confidence intervals are very broad. For example, for the NNH 95% CI's are 5 to 163); this makes the treatment effect very imprecise.

IV. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

There are obvious arguments that adult patients have more co-morbidities than pediatric patients and that in a significant portion of the adult population, delaying reintubation may bring about cardiovascular decompensation. It would have been interesting to see if the cause of death had been myocardial infarction versus ARDS or aspiration pneumonia. One other recent single-center randomized controlled trial including 81 adult patients has concluded that use of noninvasive ventilation does not prevent the need for reintubation in the ICU, but it was not powered to detect a significant difference in mortality. (1)

There is conflicting data regarding the role of noninvasive ventilation in avoiding the need for primary intubation, particularly among specific types of patients. In a previous study done on patients presenting in acute hypoxemic respiratory distress, defined as dyspnea at rest (RR > 35); a ratio of paO2:FiO2 < 200 and accessory muscle use. (2) These investigators found that complications were lower in the noninvasive group, but that patients with higher simple applied physiology scores did better with conventional ventilation. A study by Hilbert et al. with immunocompromised patients using early noninvasive positive pressure randomized versus supplemental oxygen, found a significantly lower rate of intubation and death when using early noninvasive positive pressure ventilation. (3)

Esteban's paper suggests that noninvasive ventilation may cause harm to an unselected group of adult patients who develop respiratory failure after extubation. This paper suggests that noninvasive ventilation is not a benign intervention without risks. Further studies, especially within our pediatric population, need to be conducted in order to identify subgroups of patients most likely to benefit from the application of noninvasive ventilation. However, it is important paper for our awareness of the potential harm of noninvasive ventilation given this modality's rapid rise in utilization in many PICU's.

2. Were all clinically important outcomes considered?

Death and reintubation are the two major clinically important outcomes. It may have been interesting to add complications like myocardial infarction, aspiration, pneumonitis, and difficult intubation as outcomes, but the primary objectives of death and reintubation are far and away the most important.

3. Are the likely treatment benefits worth the potential harms and costs?

According to this paper, the answer is no since the intervention appeared to be harmful.

References:

1. Keenan SP, Powers C, McCormack DG, Block G. Noninvasive positive-pressure ventilation for postextubation respiratory distress: a randomized controlled trial. JAMA 2002; 287: 3238-3244. [abstract]
2. Antonelli M, Conti G, Rocco M, Bufi M, De Blasi RA, Vivino G, Gasparetto A, Meduri GU. A Comparison of Noninvasive Positive-Pressure Ventilation in Patients with Acute Respiratory Failure. NEJM 1998; 339: 429-435. [abstract]
3. Hilbert G, Gruson D, Vargas F, Valentino R, Gbikpi-Benissan G, Dupon M, Reiffers J, Cardinaud JP. Noninvasive Ventilation in Immunosuppressed Patients with Pulmonary Infiltrates, Fever, and Acute Respiratory Failure. NEJM 2001; 344: 481-487. [abstract]

Comments

Back to the EB Journal Club Index