It is published here in its accepted form prior to copyediting and typesetting by our production team. After these production processes are complete and the authors have approved the resulting proofs, the article will move to the latest issue of the ERJ online.

MBW measurements with the new setup were feasible in infants. However, despite attempts to correct software settings, outcomes between setups were not interchangeable. Further work is needed before widespread application of the new setup can be recommended.

Unexpected Results in Infant Multiple Breath Washout: Is the Truth Hidden in the Washin?Recently, a refined software algorithm for infant multiple breath washout (MBW) analysis was proposed aiming to improve the accuracy of tracer gas evaluation. 1 The use of this algorithm revealed high lung clearance index (LCI) values in infants with cystic fibrosis (CF) which were previously found to be normal. Foong et al. used this algorithm to retrospectively analyze infant MBW data from healthy, CF, and preterm infants 2 and report an increase in LCI in CF and preterm groups. However, in the majority of healthy subjects they also found higher LCI values compared with those reported previously for healthy infants, measured either by massspectrometer 3 or by ultrasonic flowmeter systems. 1 Those high LCI values of up to 12 lung turnovers in healthy subjects are clearly counterintuitive.Besides algorithm-related errors, 4 other factors can affect MBW indices, for example, insufficient foreign tracer gas washin. In MBW tests using foreign tracer gases (in this case, SF 6 ) the tracer gas needs to equilibrate between inspired and expired tracer gas fractions before starting the washout. In the present setup for infant MBW, the operator manually stops the washin and starts the washout, but there is no clear online display of a complete washin. 1 Thus, it is easily possible to terminate the washin earlier than required. We hypothesized that a noncomplete tracer gas washin phase could lead to false washout outcomes.To examine this, we simulated different degrees of incomplete washin using a 4% SF 6 MBW measurement from a healthy infant. We truncated the washin stepwise and conventionally analyzed the washout (see online supplement, OLS). Washin incompleteness resulted in increased LCI, functional residual capacity (FRC) and cumulative expired volume (CEV), using the proposed (manual modus) as well as the old algorithm (automatic modus). 1 Using the manual modus, the increase was even more pronounced. Interestingly, the more incomplete the washin, the later the "target" washout breath (at 1/40th of the initial SF 6 concentration) where LCI is calculated (linear regression analysis, R 2 0.79, P < 0.001) and the higher the washout outcomes. We were able to reproduce the same results in vitro using a Plexiglas infant lung model (see OLS).Based on the physiology of the MBW test, during an incomplete washin the tracer gas fraction within the lungs is lower and predominantly distributed in fast ventilated lung compartments. Depending on the algorithms for determining the "target" washout breath, using a relative (1/40th) or absolute tracer gas concentration step, incomplete washin will result in shorter or longer washout times, respectively. In systems that measure molar mass (MM) and calculate those steps based on MM differences, washin incompleteness leads to overestimations in LCI, FRC, CEV ( Fig. 1). An incomplete washin leads to lower end-expiratory MM at the end of the washin (EEMM WI ) compared with the complete washin. This rela...