Respiratory oscillometry in infants: methodological aspects

<bold>Background:</bold> Oscillometry in infant lung function is of emerging potential but is challenged by special conditions (sleep, position, nasal breathing and high mechanical impedance) not encountered in older age groups. <bold>Aims:</bold> To investigate the effects of face mask positioning and head support on respiratory impedance (Zrs) in infants. <bold>Methods:</bold> Sedated infants (n=21) who were born preterm (25-37 wk gestational age) were evaluated longitudinally as outpatients at corrected-ages of 5-6 (visit 1; V1) and 12 months (V2). A device in development (N-100; Thorasys Inc., Montreal, CA) was employed to measure Zrs in the 7-41-Hz range. The head/mask support was accomplished with 1 hand (1H) or 2 hands (2H); >3 regular segments of 7 s were collected in each condition. The Zrs spectra were processed to correct (corr) for the face mask shunt. Mean resistance (R), compliance (C), and resonance frequency (fres) were estimated via model fitting. <bold>Results:</bold> Uncorrected Zrs failed to exhibit fres in 12 (V1) and 1 (V2) subjects and resulted in unexpectedly low C values. V1 and V2 results reflected growth (Table 1). There was no significant difference in any measure with 1H and 2H support (signed rank test). <table-wrap><object-id>erj;64/suppl_68/PA1462/TB1</object-id><object-id>T1</object-id><object-id>TB1</object-id><table><colgroup><col></col><col></col><col></col><col></col><col></col></colgroup><tbody><tr><td></td><td>V1</td><td>V1corr</td><td>V2</td><td>V2corr</td></tr><tr><td>R (cmH₂O.s/L)</td><td>31.9 [27.2;36.3]</td><td>33.4 [28.9;39.2]*</td><td>25.7 [19.8;29.8]^{# †}</td><td>23.2 [19.7;27.5]* ^††</td></tr><tr><td>C (mL/cmH₂O)</td><td>1.08 [0.86;1.32]</td><td>1.56 [1.13;1.90]**</td><td>1.62 [1.23;2.10]^{# †}</td><td>1.89 [1.56;2.84]** ^†</td></tr><tr><td>fres (Hz)</td><td>53.1 [39.3;61.0]</td><td>17.6 [16.1;23.6]*</td><td>28.6 [23.1;37.8]^#</td><td>16.4 [13.4;18.0]** ^†</td></tr></tbody></table></table-wrap> Table 1. Median [25%;75%] values of Zrs measures. <italic>p</italic> corr <italic>vs</italic> uncorr: <0.05*, <0.001**; V2 <italic>vs</italic> V1: <0.05^†, <0.001^†† <bold>Conclusion:</bold> Correction for the face mask shunt impedance provided more realistic Zrs spectra and derived measures, especially at high Zrs values.