Both pressure- and volume-limited non-invasive positive pressure ventilation (NPPV) have been used in patients with chronic respiratory failure. The aim of the present study was to compare the efficacy of ventilation during nocturnal volume- and pressure-limited NPPV. Fifteen patients (nine COPD, six non-COPD) were randomly assigned to receive either volume-limited or pressure-limited NPPV and were switched to the complementary mode after 6 weeks. Ten patients (five COPD, five non-COPD) completed the study. PaCO2 during sleep comparably decreased from 54.6+/-8.0 to 46.2+/-6.1 mmHg during volume-limited NPPV (P<0.05), and to 46.5+/-6.4 mmHg during pressure-limited NPPV (P<0.05). Improvements in sleep quality assessed by polysomnography were comparable, but less gastrointestinal side effects were reported for pressure-limited NPPV (P<0.05). Using a pneumotachograph the variance of inspiratory volumes was lower, but the variance of peak inspiratory pressures was higher during volume-limited NPPV compared to pressure-limited NPPV. Substantial leak volumes which accounted for 57% (volume-limited NPPV) and for 58% (pressure-limited NPPV) of the applied inspiratory volume were independent from the mode of ventilation. In conclusion, nocturnal volume- and pressure-limited NPPV have similar effects on gas exchange and sleep quality in patients with hypercapnic chronic respiratory failure, but volume-limited NPPV is associated with more gastrointestinal side effects.