Stephen J.LurieMD, PhD, Senior Editor
In Reply: We share the concern of Drs Girault and Auriant about the discrepancy in results among studies of NIV, not only in the setting of postextubation respiratory distress but in most settings. Discordant study results generally arise as a result of differences in either the population studied, the application of the technology or treatment, or the way outcomes are defined. We believe the first 2 may be most important in this instance. Our patient population was heterogeneous, representing the diverse population found in ICUs. The one randomized trial to which Girault and Auriant alluded primarily include patients after having undergone lung resection.1 This group represents patients with obstructive lung disease, a population that the literature has previously suggested strongly benefit from NIV. We specifically excluded such patients after the first year, resulting in fewer than 25% of patients with any history of obstructive lung disease and only 10% for which this was a factor in ICU admission.
Keenan SP. Differences in Success Rates of Noninvasive Ventilation—Reply. JAMA. 2002;288(20):2540. doi:10.1001/jama.288.20.2540-JLT1127-2-2