Determination of functional residual capacity (FRC) by multibreath nitrogen washout in a lung model and in mechanically ventilated patients. Accuracy depends on continuous dynamic compensation for changes of gas sampling delay time.

OBJECTIVE

Validation of an open-circuit multibreath nitrogen washout technique (MBNW) for measurement of functional residual capacity (FRC). The accuracy of FRC measurement with and without continuous viscosity correction of mass spectrometer delay time (TD) relative to gas flow signal and the influence of baseline FIO2 was investigated.

DESIGN

Laboratory study and measurements in mechanically ventilated patients.

SETTING

Experimental laboratory and anesthesiological intensive care unit of a university hospital.

PATIENTS

16 postoperative patients with normal pulmonary function (NORM), 8 patients with acute lung injury (ALI) and 6 patients with chronic obstructive pulmonary disease (COPD) were included.

INTERVENTIONS

Change of FIO2 from baseline to 1.0.

MEASUREMENTS AND MAIN RESULTS

FRC was determined by MBNW using continuous viscosity correction of TD(TDdyn), a constant TD based on the viscosity of a calibration gas mixture (TD0) and a constant TD referring to the mean viscosity between onset and end of MBNW (TDmean). Using TDdyn, the mean deviation between 15 measurements of three different lung model FRCs (FRCmeasured) and absolute volumes (FRCmodel) was 0.2%. For baseline FIO2 ranging from 0.21 to 0.8, the mean deviation between FRCmeasured and FRCmodel was -0.8%. However, depending on baseline FIO2, the calculation of FRC using TDmean and TD0 increased the mean deviation between FRCmeasured and FRCmodel to 2-4% and 8-12%, respectively. In patients (n = 30) the average repeatability coefficient was 6.0%. FRC determinations with TDmean and TD0 were 0.8-13.3% and 4.2-23.9% (median 2.7% and 8.7%) smaller than those calculated with TDdyn.

CONCLUSION

A dynamic viscosity correction of TD improves the accuracy of FRC determinations by MBNW considerably, when gas concentrations are measured in a sidestream. If dynamic TD correction cannot be performed, the use of constant TDmean might be suitable. However, in patient measurements this can cause an FRC underestimation of up to 13%.