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

A comparison of noninvasive positive-pressure ventilation and conventional mechanical ventilation in patients with acute respiratory failure.

Antonelli M, Conti G, Rocco M, Bufi M, De Blasi RA, Vivino G, Gasparetto A, Meduri GU

N Engl J Med 1998;339:429-435. [abstract]

Reviewed by Girish Deshpande, MD, Kathleen Meert, MD, Children's Hospital of Michigan

Review posted October 6, 1998

---

I. What is being studied?:

The study objective:

To compare the efficacy of noninvasive ventilation (NIV) delivered through a face mask with the efficacy of conventional mechanical ventilation (CMV) delivered through an endotracheal tube in patients with severe hypoxemia whose condition has not improved with aggressive medical therapy and who required mechanical ventilation.

The study design:

Prospective randomized study.

The patients included:

Consecutive adults with acute hypoxemic respiratory failure who were admitted to the ICU of La Sapienza University Hospital in Rome were enrolled.

Eligibility Criteria:

1. acute respiratory distress with deterioration despite aggressive medical management, including severe dyspnea at rest as determined by a clinician who was not an investigator.
2. A PaO2/ FiO2 ratio of less than 200 while the patient was breathing oxygen through a Venturi mask.
3. Active contraction of the accessory muscles of respiration or paradoxical abdominal motion.

The patients excluded:

1. COPD
2. Immunosuppressive therapy
3. Requirement of emergency intubation for CPR, respiratory arrest, severe hemodynamic instability or encephalopathy.
4. Respiratory failure caused by neurologic disease or status asthmaticus.
5. More than two new organ failures (e.g. the simultaneous presence of renal and cardiovascular failure.
6. Tracheostomy, facial deformities, recent oral, esophageal or gastric surgery.

The interventions compared:

The noninvasive ventilation group consisted of 32 patients, who were connected to the ventilator with conventional tubing to a clear, full-face mask. After the mask was secured the pressure support was increased to achieve an exhaled tidal volume of 8 to 10 ml/kg, a RR of fewer than 25 breaths per minute, the disappearance of accessory-muscle activity, and patient comfort. Continuous positive airway pressure was increased by 2 to 3 cm of water repeatedly, up to 10 cm of water, until the FIO2 requirement was 0.6 or less.

The conventional mechanical ventilation group consisted of 32 intubated patients. Ventilator settings used were IMV (rate, 4 to 7 breaths/min) with pressure support (14 to 20 cm of water), adjusted to achieve a spontaneous Vt of 8 to 10 ml/kg, a RR of fewer than 25 breaths per minute, and the disappearance of accessory-muscle activity. PEEP was increased in increments of 2-3 cm H2O until FIO2 requirement was 0.6 or less. All patients were weaned from the ventilator by reducing the level of pressure support by 4 cm of water twice and then decreasing the rate by two breaths per minute at two-hour intervals, as tolerated. Patients who tolerated an IMV rate of 0.5 breaths/min, PS level of 8 cm of water, and an FiO2 of 0.5 or less had a two hour T-piece trial. These patients then underwent extubation if they maintained a RR lower than 30 breaths/min and PaO2 greater than 75 mm Hg.

The outcomes evaluated:

Primary end points:

1. Gas exchange: PaO2/FIO2 values were determined at baseline, at one hour (for initial improvement), at four-hour intervals (for sustained improvement) during mechanical ventilation, and before discontinuation of ventilatory support. Improvement was defined as the ability to increase the PaO2:FIO2 ratio to >200 or an increase in this ratio of >100 from base line
2. Complications of mechanical ventilation: pneumonia, sepsis and sinusitis.

Secondary end points:

1. Survival.
2. The duration of mechanical ventilation.
3. The duration of the stay in the ICU.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. The patients were randomly assigned to receive either conventional mechanical ventilation (CMV) with endotracheal intubation or NIV through a face mask. They do not mention the method of randomization.

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

Was followup complete?

Yes. All patients were followed up after randomization. Of 77 patients who met the entry criteria, 13 chose not to participate; thus 64 were enrolled. Thirty two patients were assigned to each group. Nine patients in the NIV group and 15 in the CMV group died in the ICU and 40 patients survived to be discharged from the ICU. Two patients (one in each group) died in the hospital after discharge from the ICU. One, in NIV group, died of ventricular fibrillation. The other, in the CMV group, died of cardiogenic shock due to a new myocardial infarction. The other patients were successfully discharged from the hospital without further complications.

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

Yes. Ten patients in the NIV group (31%) required endotracheal intubation an average of 15 ± 7 hours after entry into the study. However these patients were analyzed in the group to which they were originally randomized. The reasons for intubations were the failure of NIV to maintain PaO2 above 65 mm Hg (4 patients), inability to correct dyspnea (1 patient), inability to manage copious secretions (1 patient), intolerance of NIV (2 patients), and hemodynamic instability (2 patients).

Criteria for switching from NIV to CMV (NIV failure):

1. failure to maintain a PaO2 above 65 mm Hg with an FIO2 of at least 0.6.
2. the development of the condition necessitating endotracheal intubation to protect the airways (coma or seizure disorder).
3. to manage copious tracheal secretions.
4. hemodynamic or electrocardiographic instability
5. an inability on the part of the patient to tolerate the face mask because of discomfort.

Secondary questions:

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

No. Investigators were not blind to the use of NIV or CMV.

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

Yes. The two groups were similar in most clinical characteristics (e.g. age, sex, Simplified Acute Physiologic Score, HR, Systolic BP, and causes of respiratory failure), except for arterial pH and PaCO2. In CMV group the mean arterial pH was significantly lower (7.37 ± 0.1 vs. 7.45 ± 0.1 ) and more patients had a PaCO2 > 45 mm Hg (12 vs. 5 patients).

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

The treatments for the conditions precipitating respiratory failure and the ventilatory management was similar. The mean level of applied PEEP was similar between the two groups. However, the investigators do not provide any detail regarding other aspects of management. It is difficult to be sure that the additional medical management of both groups was similar.

III. What were the results?

1. How large was the treatment effect?

a) Risk of serious complications: Twelve of 32 patients in the NIV group and 21 of 32 patients in the CMV group had serious complications. Entering this data in a 2 X 2 table as shown below, we calculated the relative risk (RR) of 0.57. RR value < 1 indicates a lower risk of complications with NIV.

	Serious complications
	Present	Absent	Totals
NIV group	12	20	32
CMV group	21	11	32
Totals	33	31	64

The absolute risk reduction (ARR) is 0.28 or 28% [The proportion with complications in conventional group (X= 0.66) minus the proportion with complications in NIV group (Y= 0.38); 0.66-0.38]. The relative risk reduction (RRR) is 42.5% [RRR = (1-(Y/X)) x 100%]. The number needed to treat (NNT= 1/ARR) is 3.5. This represents the number needed to treat to prevent a complication in one patient.

b) Number of deaths: Nine patients (28%) in the NIV group and 15 (47%) in the CMV group died in the ICU.

	Survival and deaths
	Died	Survived	Totals
NIV group	9	23	32
CMV group	15	17	32
Totals	24	40	64

By using the method of analysis similar to above we calculated a RR of 0.6, ARR of 0.19 or 19%, RRR of 40%, for the risk of death. (Because the number of deaths was not statistically significant and the confidence interval for RR crosses 1, we did not calculate NNT).

c) Sustained improvement in oxygenation: The oxygenation improved over time in 22 patients in the NIV group and in 17 patients in the CMV group.

	Sustained Improvement in Oxygenation
	Present	Absent	Totals
NIV group	22	10	32
CMV group	17	15	32
Totals	39	25	64

We calculated a RR of 1.28 [relative risk of more than 1 indicates increase "risk" of sustained improvement in oxygenation (i.e., a benefit of NIV)], ARR of -0.15 or -15% (indicates an absolute "risk" increase for sustained improvement with use of NIV), RRR of -0.28 or -28% (or relative "risk" increase by 28% for improvement in oxygenation.

d) Length of ICU stay: Was shorter for patients in the NIV group compared to CMV group (9 ± 7 days vs. 16 ± 17 days in the CMV group, p=0.04).

e) Duration of mechanical ventilation: Among 40 patients who survived to be discharged from the ICU, the patients in the NIV group had a shorter duration of mechanical ventilation (3 ± 3 days vs. 6 ± 5 days, p=0.006)

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

An estimate of the treatment effect is reflected in the confidence intervals for RR, ARR, RRR, NNT. The confidence intervals are very wide for all the parameters.

a) Risk of serious complications:

	95% Confidence Intervals
RR	34 to 95%
RRR	5 to 66%
ARR	4.7 to 51.3%
NNT	2 to 21

b) Risk of Death:

	95% Confidence Intervals
RR	30 to 115%
ARR	-4.5 to 42.5%
RRR	-15 to 69%

c) Sustained improvement in oxygenation:

	95% Confidence Intervals
RR	85 to 190%
ARR	-38.52 to 8.52%
RRR	-90 to 15%

The negative values for the lower confidence limits of ARR and RRR represent risk increase. When the confidence intervals for ARR and RRR include zero, ARR and RRR are not significant. For example, when ARR is negative, its inverse represents the number needed to harm rather than the number needed to treat. Therefore, NNT is only calculated for the effects where the 95% confidence interval around the RR does not cross 1.

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

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

This study included only adults. The difficulty in applying these results to pediatric patients is that a) the disease processes causing respiratory failure are different in adults than children, b) there is not sufficient data in the pediatric age group about the use of NIV. The data suggests that patients in the NIV group have initial and sustained improvement in oxygenation and less complications compared to CMV group. Keep in mind however, complications such as nosocomial pneumonia are difficult to diagnose in patients with pre-existing respiratory failure. Criteria for diagnosing sinusitis are not clearly defined in the paper. Because the study is not blinded, there may be bias in identifying the complications in each patient. It is not clear from the paper whether NIV or CMV were started immediately after randomization. If the CMV group had to wait longer than the NIV group to receive mechanical ventilation, CMV patients could have deteriorated further before mechanical ventilation was initiated. In addition, other problems one may come across in pediatric patients are less tolerance as compared to adults, and improper fit of the face mask due to a smaller facial structure.

2. Were all clinically important outcomes considered?

Yes. Initial and sustained improvement in oxygenation, serious complications, duration of ICU stay, and survival were all considered.

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

Yes. The NIV group was effective not only in improving oxygenation but also in reducing the number of complications, duration of ventilation and ICU stay. This should reduce the cost and increase safety.

Comments

Back to the EB Journal Club Index