© 2004 American Thoracic Society
Noninvasive Ventilation for Pulmonary Edema in the Emergency RoomTo the Editor:We read with interest the study by Nava and coworkers (1). In their trial, performed on patients with severe acute pulmonary edema, noninvasive pressure support ventilation (NIPSV) improved oxygenation and dyspnea more rapidly than conventional oxygen therapy, but it failed to decrease the intubation rate, which was only reduced in patients with hypercapnia. As the authors commented, their intubation rate with NIPSV (20%) was higher than that of our study (2) using NIPSV in patients with acute pulmonary edema (6%), especially in patients without hypercapnia (34% versus 8%). Their trial was performed in emergency rooms by teams with limited experience in NIPSV and using very simple portable ventilators, not equipped with tidal volume display. The inspiratory support was 14.3 ± 21.1 cm H2O (mean ± SD) with 6.1 ± 3.2 cm H2O of positive end-expiratory pressure (PEEP). This level of inspiratory support was between what we applied in our trial (20 cm H2O with 5 cm H2O PEEP) and that reported by Sharon and coworkers (3) (9 cm H2O with 4 cm H2O PEEP), who used ventilators with similar characteristics to those of the present study and had a very high rate of failure. Conversely, in our study we used ICU ventilators and adjusted the level of inspiratory support to the displayed tidal volume, which was 531 ± 143 ml. We often had to press the masks manually to avoid excessive leakage, and administered additional doses of morphine to improve patient compliance. In the study of Nava and colleagues (1), the inspiratory support was set to "the maximum tolerated," which could be rather subjective. The wide dispersion on the reported inspiratory support (SD > mean) suggests a high variability in the management. Considering these factors and the low doses of morphine used in this trial, it is possible that some patients were ventilated with low tidal volumes, which could have worsened the edema (4), contributing to increased intubation rate. Probably, the new generation of ventilators equipped with tidal volume display and leakage compensation would have been more appropriate in that context. Furthermore, the large list of criteria for endotracheal intubation could also have precipitated early intubations. Finally, we would like to point out that in a recent observational study (5), we found that patients with hypercapnia or severe acidosis (pH < 7.25) not showing severe hypertension (> 180 mm Hg) had a high intubation rate (> 50%) when they were managed with conventional therapy. Therefore, the significant benefit of NIPSV in patients with hypercapnia demonstrated by Nava and coworkers (1) opens the door to definitely considering NIPSV as a first choice for treatment in these patients.
Hospital Dos de Maig, CSI University of Barcelona Barcelona, Spain FOOTNOTES Conflict of Interest Statement: J.M., J.P., A.J.B., and F.V. have no declared conflict of interest. REFERENCES
From the Authors: I thank Dr. Masip and colleagues for their interest and useful comments. First, it is true that the "limited" experience of the teams involved may have influenced the outcome of our patients (1). For example, we have recently shown how the different attitudes of the involved staff regarding noninvasive pressure support ventilation (NIPSV) may vary with time, influencing the success rate of NIPSV (2). Although most of the studies using noninvasive ventilation in the treatment of acute pulmonary edema were performed in single centers, usually in the ICU, our investigation is the first to have been performed in the emergency departments of five different hospitals. This gave us the unique opportunity to focus our attention on the "real world," in which physicians are not specifically trained for NIPSV. As demonstrated for hypercapnic respiratory failure caused by chronic obstructive pulmonary disease (COPD) (3), early treatment with NIPSV outside the ICU may result in higher intubation and death rates in patients with more severe symptoms compared with other studies performed in "more protected" environments. Second, we have used a portable ventilator for delivering NIPSV. This kind of ventilator has the clear advantage of being easier to use, less noisy, and lighter than most of the ICU ventilators. Unfortunately, most portable ventilators have a limited monitoring system so that tidal volume was not recorded. Massive air leaks were not detected "clinically" in three patients who nevertheless did not need intubation. It is worth noting that the ventilator we have used gives us the possibility of compensating for air leaks, whereas most of the ICU ventilators do not. Lastly, the inspiratory pressures that we have applied are in keeping with most of the other studies performed in acute respiratory failure (4). Third, concerning the use of morphine, in the absence of definitive guidelines, we were rather "conservative" (maximum 4 mg); but as illustrated in Table 3 of our article (1), only three patients needed to stop NIPSV for intolerance, so that the use of higher doses of morphine would not have changed our overall results. In summary, we think that the main difference between our study (1) and that of Masip and coworkers (5) reflects the difference between an environment that is "protected," experienced, and familiar with the use of NIPSV and the "real life" situation that most hospitals have to face daily. We agree nevertheless that the significant benefit of NIPSV in patients with hypercapnia opens the door to consider noninvasive ventilation as the "first-line treatment" in this subset of patients.
Fondazione S. Maugeri Pavia, Italy FOOTNOTES Conflict of Interest Statement: S.N. has no declared conflict of interest. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
