Noninvasive Ventilation
http://www.100md.com
《新英格兰医药杂志》
To the Editor: The study by Esteban and colleagues (June 10 issue)1 was stopped on the basis of an interim analysis after 43 deaths. A stopping rule based on a P value of 0.048 was invoked (stopping boundary, P=0.05). Suppose the study had continued, with further monitoring by the data safety and monitoring committee, and five extra patients had been recruited into each group. If another death had occurred with standard care, the P value would have been greater than 0.07. If another death had occurred in each group, the P value would have been greater than 0.05. In either instance, the study would have continued. This scenario illustrates the instability of the P value around the stopping boundary.
In the First International Study of Infarct Survival (ISIS-1), there appeared to be excess mortality with beta-blockers after 120 deaths had occurred (P<0.05).2 Nevertheless, the trial continued and, after 700 deaths, there was a significant reduction in mortality3 — a finding that was supported by the results of other trials.2 If Esteban and colleagues had adopted a more stringent stopping rule,4 one that accounted for all the available data,5 the study might have continued and provided more reliable evidence. Now, further trials are still necessary.
William O. Tarnow-Mordi, M.R.C.P., F.R.C.P.C.H.
University of Sydney
Sydney, NSW 2145, Australia
williamt@westgate.wh.usyd.edu.au
Val Gebski, M.Stat.
University of Sydney
Sydney, NSW 2050, Australia
Anne Cust, M.P.H.
National Health and Medical Research Council
Clinical Trials Centre
Sydney, NSW 2050, Australia
References
Esteban A, Frutos-Vivar F, Ferguson ND, et al. Noninvasive positive-pressure ventilation for respiratory failure after extubation. N Engl J Med 2004;350:2452-2460.
Yusuf S. Challenges in the conduct and interpretation of phase II (pilot) randomized trials. Am Heart J 2000;139:S136-S142.
First International Study of Infarct Survival Collaborative Group. Randomised trial of intravenous atenolol among 16 027 cases of suspected acute myocardial infarction: ISIS-1. Lancet 1986;2:57-66.
Peto R, Pike MC, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. I. Introduction and design. Br J Cancer 1976;34:585-612.
Peter JV, Moran JL, Phillips-Hughes J, Warn D. Noninvasive ventilation in acute respiratory failure -- a meta-analysis update. Crit Care Med 2002;30:555-562.
To the Editor: We think that some aspects of the noninvasive-ventilation technique could have been responsible for the lack of effectiveness in the trial conducted by Esteban et al. The ventilator mode they chose was pressure support set to achieve a tidal volume of more than 5 ml per kilogram of body weight. This may have been insufficient, since it is usually recommended that tidal volume be set at about 10 to 15 ml per kilogram. The authors do not comment on the pressures reached in the airway. The authors titrated the fraction of inspired oxygen and the positive end-expiratory pressure to maintain the arterial oxygen saturation above 90 percent. However, end-expiratory pressure must be set to prevent rebreathing, to keep alveoli open in patients with very low residual functional capacity, and, particularly in patients with chronic obstructive pulmonary disease, to counterbalance intrinsic positive end-expiratory pressure.1 The absence of these data prevents us from knowing whether lack of effectiveness is related to noninvasive ventilation or to the method with which we perform noninvasive ventilation.2
Salvador Díaz-Lobato, Ph.D.
Hospital Universitario La Paz
28046 Madrid, Spain
sd101m@nacom.es
Sagrario Mayoralas-Alises, M.D.
Hospital de Móstoles
28935 Mostoles, Spain
José Villamor, Ph.D.
Hospital Universitario La Paz
28046 Madrid, Spain
References
Díaz Lobato S, Mayoralas-Alises S. Noninvasive ventilation. Arch Bronconeumol 2003;39:566-579.
Truwit JD, Bernard GR. Noninvasive ventilation -- don't push too hard. N Engl J Med 2004;350:2512-2515.
To the Editor: Esteban and colleagues report that noninvasive ventilation may increase the risk of death and that it does not decrease the risk of a need for reintubation in patients with respiratory failure after extubation. In their study, 28 patients in the standard-therapy group crossed over to receive rescue noninvasive ventilation. If these patients are considered to have had treatment failure, and if they are included with the other patients in that group who were reintubated, then the standard-therapy group had a significantly higher risk of a need for reintubation than the noninvasive-ventilation group (relative risk, 1.39; 95 percent confidence interval, 1.11 to 1.76; P<0.001). Other authors have used this end point when evaluating respiratory failure after extubation.1
The 28 patients who crossed over may have been the sickest patients in the standard-therapy group and may have had a higher mortality rate if not for the use of noninvasive ventilation as rescue therapy — a point that is not discussed in the article. More information regarding the baseline characteristics of these patients would be helpful. It is possible that noninvasive ventilation decreased the risk of a need for reintubation, since only 7 of the 28 patients who received rescue therapy (25 percent) required reintubation.
David A. Lipson, M.D.
Jeffrey S. Sager, M.D.
University of Pennsylvania
Philadelphia, PA 19104
dlipson@mail.med.upenn.edu
References
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.
The authors reply: We agree with the hypothetical scenario described by Tarnow-Mordi and colleagues, but when the study was conducted, there was a group with an increased mortality rate. We were required to stop the study immediately. Our study is not comparable to ISIS-1 because in that study, there were no predefined stopping criteria. The excess mortality in the experimental group after 120 deaths is an observation described after the study was published.
Lipson and Sager analyze our data on the basis of a more strict analysis according to intention to treat than the analysis we performed. They consider the 28 patients assigned to standard medical therapy who crossed over to noninvasive ventilation (including the 21 patients who did not require reintubation) as having had treatment failure. Since the baseline characteristics of these patients were not significantly different from those of the overall group, it is difficult to argue that mortality would have been higher in the noninvasive-ventilation group. It is true that the rate of reintubation in the group of patients who crossed over was significantly lower than the rate in the group assigned to noninvasive ventilation. But we think that the size of the crossover group and the decision to use noninvasive ventilation that was not by protocol in patients who had been randomly assigned to standard medical therapy make it difficult to draw clinically relevant conclusions.
We agree with Díaz-Lobato and colleagues that the success of noninvasive ventilation can be related to the method of application of the technique. However, in the study by Keenan et al.,1 specific training did not improve the results, and our study was performed in intensive care units with experience in this technique. Thus, we do not agree with the suggestion that the method we used to perform noninvasive ventilation was the factor responsible. We also cannot agree with the assertion that the tidal volume should be set to about 10 to 15 ml per kilogram. This tidal volume has been recommended for volume-limited ventilation because of the need to compensate for air leaks. Experts2,3 and participants in a consensus conference on noninvasive ventilation4 recommend starting with low pressures and volumes in spontaneously triggered mode and suggest that these settings be adjusted to provide the lowest inspiratory pressures or volumes needed to improve patients' comfort (by decreasing both the respiratory rate and respiratory-muscle unloading) and to improve gas exchange.
Andrés Esteban, M.D., Ph.D.
Fernando Frutos-Vivar, M.D.
Hospital Universitario de Getafe
28905 Madrid, Spain
aesteban@ucigetafe.com
Antonio Anzueto, M.D.
University of Texas Health Science Center
San Antonio, TX 78229-3900
References
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.
Hillberg RE, Johnson DC. Noninvasive ventilation. N Engl J Med 1997;337:1746-1752.
Mehta S, Hill NS. Noninvasive ventilation. Am J Respir Crit Care Med 2001;163:540-577.
International Consensus Conferences in Intensive Care Medicine: noninvasive positive pressure ventilation in acute respiratory failure. Am J Respir Crit Care Med 2001;163:283-291.
In the First International Study of Infarct Survival (ISIS-1), there appeared to be excess mortality with beta-blockers after 120 deaths had occurred (P<0.05).2 Nevertheless, the trial continued and, after 700 deaths, there was a significant reduction in mortality3 — a finding that was supported by the results of other trials.2 If Esteban and colleagues had adopted a more stringent stopping rule,4 one that accounted for all the available data,5 the study might have continued and provided more reliable evidence. Now, further trials are still necessary.
William O. Tarnow-Mordi, M.R.C.P., F.R.C.P.C.H.
University of Sydney
Sydney, NSW 2145, Australia
williamt@westgate.wh.usyd.edu.au
Val Gebski, M.Stat.
University of Sydney
Sydney, NSW 2050, Australia
Anne Cust, M.P.H.
National Health and Medical Research Council
Clinical Trials Centre
Sydney, NSW 2050, Australia
References
Esteban A, Frutos-Vivar F, Ferguson ND, et al. Noninvasive positive-pressure ventilation for respiratory failure after extubation. N Engl J Med 2004;350:2452-2460.
Yusuf S. Challenges in the conduct and interpretation of phase II (pilot) randomized trials. Am Heart J 2000;139:S136-S142.
First International Study of Infarct Survival Collaborative Group. Randomised trial of intravenous atenolol among 16 027 cases of suspected acute myocardial infarction: ISIS-1. Lancet 1986;2:57-66.
Peto R, Pike MC, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. I. Introduction and design. Br J Cancer 1976;34:585-612.
Peter JV, Moran JL, Phillips-Hughes J, Warn D. Noninvasive ventilation in acute respiratory failure -- a meta-analysis update. Crit Care Med 2002;30:555-562.
To the Editor: We think that some aspects of the noninvasive-ventilation technique could have been responsible for the lack of effectiveness in the trial conducted by Esteban et al. The ventilator mode they chose was pressure support set to achieve a tidal volume of more than 5 ml per kilogram of body weight. This may have been insufficient, since it is usually recommended that tidal volume be set at about 10 to 15 ml per kilogram. The authors do not comment on the pressures reached in the airway. The authors titrated the fraction of inspired oxygen and the positive end-expiratory pressure to maintain the arterial oxygen saturation above 90 percent. However, end-expiratory pressure must be set to prevent rebreathing, to keep alveoli open in patients with very low residual functional capacity, and, particularly in patients with chronic obstructive pulmonary disease, to counterbalance intrinsic positive end-expiratory pressure.1 The absence of these data prevents us from knowing whether lack of effectiveness is related to noninvasive ventilation or to the method with which we perform noninvasive ventilation.2
Salvador Díaz-Lobato, Ph.D.
Hospital Universitario La Paz
28046 Madrid, Spain
sd101m@nacom.es
Sagrario Mayoralas-Alises, M.D.
Hospital de Móstoles
28935 Mostoles, Spain
José Villamor, Ph.D.
Hospital Universitario La Paz
28046 Madrid, Spain
References
Díaz Lobato S, Mayoralas-Alises S. Noninvasive ventilation. Arch Bronconeumol 2003;39:566-579.
Truwit JD, Bernard GR. Noninvasive ventilation -- don't push too hard. N Engl J Med 2004;350:2512-2515.
To the Editor: Esteban and colleagues report that noninvasive ventilation may increase the risk of death and that it does not decrease the risk of a need for reintubation in patients with respiratory failure after extubation. In their study, 28 patients in the standard-therapy group crossed over to receive rescue noninvasive ventilation. If these patients are considered to have had treatment failure, and if they are included with the other patients in that group who were reintubated, then the standard-therapy group had a significantly higher risk of a need for reintubation than the noninvasive-ventilation group (relative risk, 1.39; 95 percent confidence interval, 1.11 to 1.76; P<0.001). Other authors have used this end point when evaluating respiratory failure after extubation.1
The 28 patients who crossed over may have been the sickest patients in the standard-therapy group and may have had a higher mortality rate if not for the use of noninvasive ventilation as rescue therapy — a point that is not discussed in the article. More information regarding the baseline characteristics of these patients would be helpful. It is possible that noninvasive ventilation decreased the risk of a need for reintubation, since only 7 of the 28 patients who received rescue therapy (25 percent) required reintubation.
David A. Lipson, M.D.
Jeffrey S. Sager, M.D.
University of Pennsylvania
Philadelphia, PA 19104
dlipson@mail.med.upenn.edu
References
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.
The authors reply: We agree with the hypothetical scenario described by Tarnow-Mordi and colleagues, but when the study was conducted, there was a group with an increased mortality rate. We were required to stop the study immediately. Our study is not comparable to ISIS-1 because in that study, there were no predefined stopping criteria. The excess mortality in the experimental group after 120 deaths is an observation described after the study was published.
Lipson and Sager analyze our data on the basis of a more strict analysis according to intention to treat than the analysis we performed. They consider the 28 patients assigned to standard medical therapy who crossed over to noninvasive ventilation (including the 21 patients who did not require reintubation) as having had treatment failure. Since the baseline characteristics of these patients were not significantly different from those of the overall group, it is difficult to argue that mortality would have been higher in the noninvasive-ventilation group. It is true that the rate of reintubation in the group of patients who crossed over was significantly lower than the rate in the group assigned to noninvasive ventilation. But we think that the size of the crossover group and the decision to use noninvasive ventilation that was not by protocol in patients who had been randomly assigned to standard medical therapy make it difficult to draw clinically relevant conclusions.
We agree with Díaz-Lobato and colleagues that the success of noninvasive ventilation can be related to the method of application of the technique. However, in the study by Keenan et al.,1 specific training did not improve the results, and our study was performed in intensive care units with experience in this technique. Thus, we do not agree with the suggestion that the method we used to perform noninvasive ventilation was the factor responsible. We also cannot agree with the assertion that the tidal volume should be set to about 10 to 15 ml per kilogram. This tidal volume has been recommended for volume-limited ventilation because of the need to compensate for air leaks. Experts2,3 and participants in a consensus conference on noninvasive ventilation4 recommend starting with low pressures and volumes in spontaneously triggered mode and suggest that these settings be adjusted to provide the lowest inspiratory pressures or volumes needed to improve patients' comfort (by decreasing both the respiratory rate and respiratory-muscle unloading) and to improve gas exchange.
Andrés Esteban, M.D., Ph.D.
Fernando Frutos-Vivar, M.D.
Hospital Universitario de Getafe
28905 Madrid, Spain
aesteban@ucigetafe.com
Antonio Anzueto, M.D.
University of Texas Health Science Center
San Antonio, TX 78229-3900
References
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.
Hillberg RE, Johnson DC. Noninvasive ventilation. N Engl J Med 1997;337:1746-1752.
Mehta S, Hill NS. Noninvasive ventilation. Am J Respir Crit Care Med 2001;163:540-577.
International Consensus Conferences in Intensive Care Medicine: noninvasive positive pressure ventilation in acute respiratory failure. Am J Respir Crit Care Med 2001;163:283-291.