Document resource
Introduction Multiple-breath nitrogen washout (N2 MBW) is increasingly used to derive functional residual capacity (FRC) and lung clearance index (LCI), yet device comparability remains a barrier to interchangeability in clinical and research settings. We compared FRC and LCI between the Exhalyzer (EcoMedics) and Vyntus One (Jaeger), evaluated within-device repeatability using coefficients of variation (CV), and compared MBW-derived FRC against body plethysmography (FRCpleth) as a reference.Methods Eleven healthy physiologists attempted 3–5 MBW trials per device in randomised order. FRCpleth was then measured, followed by spirometry. Per-subject mean values were analysed. Agreement between devices (Exhalyzer − Vyntus) was assessed using Bland–Altman (BA) bias and 95% limits of agreement (LoA), with regression of differences on magnitude to assess proportional bias. Reference-method BA for FRC used FRCpleth on the x-axis. Because clinical acceptability was prespecified as ±5%, ratio (log scale) BA was also used to express agreement as MBW/FRCpleth with ratio LoA. CVs for FRC and LCI were compared between devices using paired log ratio analysis, reporting geometric mean CV ratios (Exhalyzer/Vyntus).Results For between-device agreement, FRC bias was 0.01 L with LoA −0.57 to 0.59 L; regression suggested proportional bias (p=0.046). For LCI, Exhalyzer was lower than Vyntus (bias −1.48 units; LoA −2.92 to −0.04) without evidence of proportional bias (p=0.592). Versus FRCpleth, MBW underestimated FRC on average (Exhalyzer/FRCpleth ratio 0.885; ratio LoA 0.656–1.195; Vyntus/FRCpleth ratio 0.900; ratio LoA 0.737–1.098), exceeding the ±5% criterion (0.95–1.05). Repeatability differed for FRC (CV ratio 1.56; 95% CI 1.06–2.28; p=0.027), but not for LCI (CV ratio 0.99; p=0.985). Table 1Conclusion In healthy subjects, Exhalyzer and Vyntus were not interchangeable for LCI and showed limited agreement for FRC, with evidence of magnitude dependent bias for FRC. Both MBW devices underestimated FRCpleth beyond a prespecified ±5% acceptability margin suggesting systematic differences in MBW derived lung volumes compared with the reference method. Exhalyzer demonstrated poorer repeatability for FRC than Vyntus, while LCI repeatability was comparable. These findings support caution when pooling or comparing MBW outcomes across devices and highlight the need for device specific reference ranges and/or validated harmonisation approaches when switching platforms.Abstract P45 Table 1Summary of agreement and repeatability outcomes