In a study conducted by Conti et al. [9], the effects of sufentanil on respiratory drive, respiratory pattern, and gas exchange were investigated in intubated patients during pressure support ventilation. In this study, no changes in hemodynamic or respiratory parameters were observed with the use of 0.2–0.3 μg/kg/h sufentanil infusion for 24 h. Newer drugs have a better safety profile than sufentanil and are used in ICUs for sedation and analgesia in intubated patients.

There are several recent studies, where remifentanil and dexmedetomidine is used to facilitate NPPV in patients whom refuse this treatment due to intolerance. Both drugs have shorter half-life and lower risk of unwanted effects.

Rocco et al. [10] reported 61 % of 36 patients with acute hypoxemic failure, who refused NPPV, were able to receive the treatment after sedation with remifentanil (0.025 μg/kg/min). The mortality rate in the successfully ventilated patient group was 14 %, whereas it was 50 % in the patients who required ETI despite remifentanil infusion. In a pilot study conducted on 12 patients who refused NPPV, only 4 needed ETI with the infusion of remifentanil [11]. The authors concluded that the use of remifentanil sedation was safe and effective in patients who refused noninvasive ventilation.

Huang et al. [12] compared dexmedetomidine and midazolam in NPPV failure due to patient refusal in a randomized study. Of the 62 treated patients with cardiogenic pulmonary edema, 20 had failed NPPV despite sedative infusions. The failure rate was 44.8 % in the midazolam group and 21.2 % in the dexmedetomidine group. The reasons for NPPV failure were inability to increase PaO₂/FiO₂ ratio, worsening hemodynamic status, and inability to cope with secretions. In patients with successful NPPV, dexmedetomidine resulted in shorter ICU stay and lower rates of respiratory infections, although the success rate and mortality were similar in both groups.

In a randomized, double-blind, controlled trial, dexmedetomidine was compared with placebo in patients receiving NPPV for ARF [13]. The patients were randomized to study groups within 8 h of NPPV, and received infusions for 72 h or until intubation or weaning. The study failed to show any advantage of initiating dexmedetomidine with noninvasive ventilation on tolerance, the patients were not selected from the ones who had intolerance to interface or agitation.

Propofol is widely used for conscious sedation during diagnostic interventions. It provides an easily controllable level of sedation with fast recovery after cessation of drug application. Clouzeau et al. [14] used propofol during fiber-optic bronchoscopy in patients with acute hypoxemic respiratory failure with noninvasive pressure support. The authors reported that an infusion regimen targeting plasma propofol concentrations was safe and effective in relieving patient discomfort without any adverse effects.

All of the above publications were reports of the results of preliminary or pilot studies with limited numbers of patients, generally with intolerance to NPPV. Although not conclusive, there are signs of a beneficial effect of analgosedation to relieve discomfort and improve patient acceptance of noninvasive ventilation. An important aspect of these studies is that all of them were conducted in ICUs, where all means of monitoring are available. The issue of whether sedation can be used in settings other than ICUs is controversial, and it may be dangerous because patients cannot be monitored as closely as in the ICU environment. It would be reasonable to admit the patients who refuse NPPV to the ICU to provide sedation before initiating invasive ventilation (i.e., endotracheal intubation).