Operator performance during the expiratory phase of manual hyperinflation appears to vary between physiotherapists for Mapleson-B or C circuits. Some physiotherapists release the valve but maintain compression of the bag, whereas others release both the valve and the bag. The effect of this difference on peak expiratory flow rate (PEFR) has not been reported. The aim of this study was to document the effect of maintaining bag compression during expiration on PEFR and inspiratory to expiratory flow rate ratio (I:E). Six physiotherapists with experience using manual hyperinflation participated. A within-subjects repeated measures design was used. Subjects performed manual hyperinflation using a Mapleson-C circuit with 'rapid release', releasing the valve only, or releasing both the bag and the valve, during expiration in a test lung model. Inspiratory time was controlled using a metronome and flows were measured with a heated pneumotachometer. Maintaining bag compression significantly reduced PEFR (1.54 (0.08) vs 2.00 (0.07) l/sec, p = 0.008) and increased I:E flow rate ratio (0.65 (0.04) vs 0.50 (0.02), p = 0.02) for the Mapleson-C circuit at a 1.4 litre target volume. There were no significant differences for these measures between techniques when subjects emptied the bag. The effect needs to be confirmed in the clinical setting.