Noninvasive Mechanical Ventilation in Pneumonia: Risks of Delayed Intubation. Is This “Red Line” Well Defined and Understood?
Randomized controlled trials (RCTs) describing whether NIV is beneficial in severe pneumonia are still scarce and uncommon practice in real word. Most have considered very heterogeneous populations of patients with varying causes of acute respiratory failure (ARF) including, even if a small percentage, pneumonia (2, 3). Although the application of NIV for pneumonia increased by 50% in last years, its role in pneumonia is even now controversial.
As reported by Valley et al (1), the use of NIV among marginal patients with severe pneumonia showed a not significant difference in 30-day mortality when compared with IMV (54% vs 55%). However, some aspects of this study require clarification in order to understand the real value of the results:
First, patients were assigned as their first mechanical ventilation procedure code (NIV or IMV) although then they underwent to IMV after a few hours (19% of NIV patients). In these cases, the 30-day mortality is hardly assignable to the respective treatments (NIV or IMV) without a control group. There is not a clear interpretation of these results for proper clinical practice recommendations.
Second, the main strength of the study is referred to the large population and large number of pneumonias. However, the unavailability of clinical score does not allow for identification of subgroups at highest risk. This could offer to clinicians more appropriate screening and stratification of severe pneumonia and NIV indications.
Currently, the role of NIV for severe pneumonia has been reported in the literature (4–6) with different response in various NIV criteria, ventilator setup and mixed subgroups: in patients without previous cardiac or pulmonary disease, defined “de novo,” the use of NIV was associated with a high likelihood of NIV failure and consequently high intubation rates (5), whereas, in patients with cardiac or pulmonary diseases (i.e., chronic obstructive pulmonary disease), defined “comorbidities” group, NIV success is more frequent compared with those with “de novo” ARF (74% vs 54%, respectively) (5); in patients with an impaired immune system, defined “immunodepressed” group, NIV approach has the potential to avoid endotracheal intubation and its complications (2).
Third, another question about the study (1) is the significant difference in primary diagnosis (pneumonia) for “NIV first” group and “IMV first” group (35.2% vs 16.9%, respectively). One of the major issues in severe pneumonia is the concern that increased risk is directly related to postponing intubation after an ineffective NIV trial (5). Increased duration of NIV before intubation for NIV failure is associated with increased in-hospital mortality. Variables independently associated with NIV failure in patients with pneumonia are a higher Simplified Acute Physiology Score II, a worsening radiological infiltrate at 24 hours after admission and a higher heart rate, lower PaO2/FIO2 ratio, and lower plasmatic level of bicarbonate after 1 hour of NIV (5, 7).
Fourth, among patients with pneumonia receiving NIV instead of IMV, a significant lower Medicare spending associated with NIV use was observed. Even though an ease of use and cost savings with NIV use for pneumonia must be weighed against the risk of increased mortality, an accurate stratification of the patients with pneumonia in RCTs could ensure greater NIV success rate and reduce mortality.