Table 5– Recommendations for offline washout software
Software transparency for
 All correction algorithms and factors applied to data (e.g. BTPS and temperature modelling)
 All algorithms used for subsequently calculated indices
 Method used to synchronise flow and inert marker gas concentration signals
 Normative data or upper limit of normal incorporated, including details of source and population characteristics (number of subjects, sex distribution, age range, ethnic group, etc.)
General recommendations
 Full availability of raw data, calibrated data and BTPS-converted data which should be saved and readily exportable in widely acceptable formats, e.g. ASCII (.txt) or .xls
Ability to assess accuracy of flow and inert gas concentration synchronisation, re-measure and manually adjust as necessary
 Ability to review tidal volume tracing to ensure correct identification of breath detection (start and end-points), and manually adjust as necessary
 Ability to review inert gas expirogram for each breath, and manual adjustment if necessary, to ensure correct estimation of
  End-tidal inert marker gas concentration
  SIII if SBW or if MBW SnIII analysis is being performed
 Ability to examine for and correct any gas-analyser drift occurring during the test. The zero calibration point may be useful as a reference for many of the gases used (N2, CO2, He and SF6) whilst 100% can be used for O2. Any correction applied should be clearly stated
 If available, monitor end-tidal CO2 values during MBW to screen for hyperventilation
FRC
 FRC is measured over all breaths of the washout, and updated after each breath, until a defined end-point in time. The end-point used for FRC determination should correspond to the end-test threshold used for ventilation inhomogeneity indices (e.g. LCI threshold)
 Exhaled inert gas volume must be corrected for re-inspired gas from the post-gas VD for each breath
 Reported FRC is that measured at the FRCgs. If other FRC values are reported, e.g. FRCao (i.e. FRCgs – pre-gas sampling point VD) these values should be described appropriately
 Report mean, sd and CoV of three technically acceptable measurements
  If only two technically acceptable measurements are available, report mean only, and state e.g. “based on two measurements alone”
  If FRC values are not within 10% of the highest FRC value, then alert the operator. Exclude FRC values which differ by >25% from the median FRC value across the three tests. Excluded tests should not be used for calculation of other MBW indices
Indices of global ventilation distribution inhomogeneity (e.g. LCI and moment ratios)
 Correct VT for external VD (see section E6.2 in the online supplementary material)
 Use appropriate corresponding FRC for calculation
 Report mean, sd and CoV of three technically acceptable measurements
  If only two technically acceptable measurements are available, report mean and % difference, and state “based on two measurements alone”
  If LCI values are more than 1.0 TO apart (highest versus lowest), then alert the operator to perform further tests
SnIII analysis (if performed)
 Calculation of SIII and SnIII
  SIII limits set to maximise the phase III used for linear regression, excluding phase II and phase IV contributions, and be manually adjustable, typically 50–95% of the expired volume in adults and 65–95% of the expired volume in children
  Manual adjustment of the SIII for breaths, where marked low frequency noise (or cardiogenic oscillations) or phase IV phenomena occur if automated estimations of SIII
  Expired inert gas concentration used for SIII normalisation (e.g. mean expired concentration or mean SIII concentration) should be clearly stated
 Acceptance criteria for breaths – identify and discard SnIII values of breaths that do not fulfil the following criteria
  Specific to tidal breathing protocols (e.g. paediatrics)
   Adequate expired volume for SnIII calculation: volume corresponding to SIII should be>50% of expiratory VT
   The expired volume should not be excessive: volume corresponding to SIII should not be>75% of expiratory VT
   Note: to try and achieve suitable breaths, an initial tidal breathing range of 10–15 mL·kg−1 can be used but may need to be adjusted for the individual patient depending on the expirogram seen
  Specific to adult protocols using VT of 1 L
   Expired volume should be >0.950 L
   Expired volume should not be >1.4 L
  A clear SIII should be identifiable. Failure to identify SIII due to the presence of artefact (e.g. breath hold, cardiogenic oscillations, cough) should prompt exclusion of that SnIII value
  When SnIII values are excluded do not discard the contribution of that breath to other indices (e.g. FRC and TO), only the SnIII value
  Tests should only contribute to overall SnIII analysis if at least two out of three of the breaths remain after SnIII breath exclusion. If >1/3 of SnIII values have been excluded due to above criteria then that entire test should be discarded
   Number of excluded SnIII values and reasons for exclusion should be reported
 Presentation of SnIII data
  Data collated from all acceptable breaths of the three technically acceptable MBW tests
  Acceptable first breath quality on all three tests for subsequent Sacin calculation
  In TO calculation, FRC and VT are calculated from the same airstream reference point used in ventilation inhomogeneity indices (see the online supplementary material section E6.2)
  Data displayed graphically as SnIII (y-axis) versus TO for each breath (x-axis)
   SnIII and SnIII × VT (i.e. VT-corrected SnIII) displayed for each breath on two separate graphs.
  These indices rely on the fact that DCDI generates a horizontal asymptote and CDI does not and are therefore only valid in cases where SnIII progression does not show a horizontal asymptote
 Clinical indices calculation
  Sacin calculation
   Requires three technically acceptable first breath SnIII values
   Sacin calculated as the mean SnIII of the three first breaths minus the Scond contribution (based on the mean TO value of the three first breaths)
  Scond calculation
   Scond calculated as the linear regression of SnIII values between approximately 1.5 and 6.0 TO
   Calculate 95% CI of the Scond regression, reject outlying values and repeat linear regression; data should be pooled from all three runs
 If SnIII analysis is performed with only two or less technically acceptable MBW tests, this should be clearly stated on the report and results interpreted with caution
 SBW SIII
  Report as mean, sd and CoV of three technically acceptable measurements
   If only two technically acceptable measurements are available, report mean and actual difference, and state “based on two measurements”
   VC measurements not within 10% of highest VC value across the SBW tests, then alert the operator
  Report both SIII (%·L−1) and SIII × expiratory VC (%) separately.
  • BTPS: body temperature, ambient pressure, saturated with water; ASCII: American Standard Code for Information Interchange; SIII: phase III slope; SBW: single-breath washout; MBW: multiple-breath washout; SnIII: normalised SIII; N2, nitrogen; CO2: carbon dioxide; He: helium; and SF6: sulfur hexafluoride; O2: oxygen; FRC: functional residual capacity; LCI: lung clearance index; VD: deadspace volume; FRCao: FRC at the airway opening; FRCgs: FRC measured at the gas sampling point; CoV: coefficient of variation; VT: tidal volume; TO: lung turnovers, calculated as cumulative expired volume/FRC; DCDI: diffusion convection-interaction-dependent inhomogeneity; Sacin: DCDI contribution to first breath SnIII. CDI: convection-dependent inhomogeneity; Scond: rate of increase of SnIII from 1.5–6 TOs, also CDI contribution to SnIII; VC: vital capacity.