The analysis of flow and pressure curves generated by ventilators can be useful in the individuation of patient-ventilator asynchrony, notably in COPD patients. To date, however, a real clinical benefit of this approach to optimize ventilator setting has not been proven. The aim of the present study was to compare: optimized ventilation, driven by the analysis of flow and pressure curves, and standard setting (same initial setting, same time at the bedside, same physician, while the ventilator screen was obscured with numerical data always available). The primary aim was the normalization of pH at two hours, whilst secondary aims were change in PaCO2, respiratory rate, patient's tolerance to ventilation (all parameter evaluated at baseline, 30, 120, 360 minutes and 24 hours after the beginning of ventilation). 70 patients (26 females, aged 78±9 years, PaCO2 74±15 mmHg, pH 7.28±0.05, mean±SD) have been enrolled, with no basal difference between the two groups.

Inclusion Criteria:

• Consecutive patients aged > 40 years affected by COPD exacerbation (defined as an acute change in a patient's baseline dyspnoea, cough and/or sputum beyond day-to-day variability sufficient to warrant a change in therapy), and respiratory acidosis (pH < 7.35) that were treated by NIV in addition to standard medical therapy

Exclusion Criteria:

• The needing of intubation
• The lack of informed consent