Acinar and conductive ventilation heterogeneity in severe CF lung disease: Back to the model

https://doi.org/10.1016/j.resp.2013.05.011Get rights and content

Highlights

  • In severe CF lung disease, interpretation of multibreath washout (MBW) is difficult.

  • Conceptual models of ventilation mechanisms can help MBW interpretation.

  • Acinar and conductive MBW indices can be linked to specific ventilation mechanisms.

  • The acinar MBW index likely reflects the best ventilated peripheral units.

Abstract

Severe convective ventilation heterogeneity occurring in CF lung disease requires a modified method to determine acinar and conductive components of ventilation heterogeneity from normalized phase III slope (Sn) curves. Modified Sacin* and Scond* (as opposed to standard Sacin and Scond) are proposed and interpreted on the basis of 2 conceptual mechanisms: (a) flow asynchrony between two convection-dependent units with a different specific ventilation, but with an identical acinus inside each unit (generating an identical diffusion-convection-dependent portion of Sn); (b) different specific ventilation (without any flow asynchrony) between two convection-dependent units with the worst ventilated unit containing an abnormal acinus generating the greatest diffusion-convection-dependent portion of Sn. In CF patients with an abnormal lung clearance index (LCI), Scond* (but not Scond) and Sacin* were significant contributors to LCI (β(Scond*) = 0.70; β(Sacin*) = 0.49; P < 0.001 for both). Mechanism (a) can entirely account for experimental Scond* values, while mechanism (b) implies that experimental Sacin* values are likely dominated by peripheral ventilation heterogeneity in the best ventilated portions of the lung.

Introduction

The degree of ventilation heterogeneity in cystic fibrosis (CF) patients can be so large that sizeable portions of the lungs are ventilated (by convection) to a very different extent than the rest of the lungs. This can be observed from state-of-the-art MRI ventilation images obtainable in CF patients (McMahon et al., 2006) or from indirect measurement of ventilation heterogeneity at the mouth (Verbanck et al., 2012a). Inhaled and exhaled aerosols have been used to non-invasively measure the degree of convection-dependent ventilation heterogeneity in cystic fibrosis (Brown et al., 1998), but inert gas washout techniques are more widely used and potentially include both convection and diffusion dependent phenomena (Paiva and Engel, 1987). Particularly in patients with various lung diseases susceptible to early peripheral structural changes, inert gas phase III slope derived indices have been suggested to be superior diagnostic tools to spirometric indices (Verbanck, 2012), yet in advanced lung disease its role has not been fully established.

From the multiple breath washout (MBW) test, a normalized phase III slope analysis can also be performed in order to separate conductive and acinar components of ventilation heterogeneity (Scond and Sacin), as a result of convection-dependent inhomogeneity (CDI), and diffusion-convection-dependent inhomogeneity (DCDI), respectively (Crawford et al., 1985, Crawford et al., 1986, Verbanck and Paiva, 1990). However, in advanced CF lung disease, the standard derivation of Sacin and Scond from normalized phase III slopes (Sn) (Verbanck and Paiva, 2011) can become invalidated by the severity of ventilation heterogeneity in these patients. In particular, Scond which is usually quantified as the rate of rise of Sn with lung turnover (TO) up to TO = 6, assumes that Sn steadily increases as the washout progresses in this TO range. Two different research groups studying ventilation heterogeneity in CF have reported that there appears to be a maximum to how abnormal Scond can become (Gustafsson, 2007, Horsley et al., 2008). This indicates that under conditions of severe ventilation heterogeneity, the rate of Sn rise up to TO = 6 starts to level off towards the theoretical horizontal Sn asymptote which is usually outside the measurement range (Verbanck and Paiva, 2011). This artificially attenuates the value of Scond even though ventilation heterogeneity becomes more severe. In the extreme case where a horizontal Sn asymptote arises during a MBW measurement range, this would even violate one of the basic assumptions for distinguishing CDI from DCDI via Scond and Sacin (Crawford et al., 1985, Verbanck and Paiva, 2011). This assumption is that DCDI (leading to Sacin) generates a horizontal Sn asymptote within the measurement range, while CDI (leading to Scond) does not.

We have previously shown that CDI can become very severe in adult CF (Verbanck et al., 2012a), based on semilog washout concentration curves alone. The associated degree of specific ventilation heterogeneity can be derived from the curvilinearity of the semilog washout curve (Curv) and from its decay most widely quantified now by the lung clearance index (LCI) (Horsley, 2009, Robinson et al., 2009, Robinson et al., 2010). LCI increases due to dead space effects and intra-acinar ventilation heterogeneity (CDI and DCDI effects) and due to increased specific ventilation heterogeneity amongst convection-dependent units (CDI effects); Curv is mainly increased by CDI effects (Verbanck et al., 2012a, Verbanck et al., 2012b). While several components of LCI are impossible to quantify, a simple 2-compartment model with no dead space can be used to mimic the experimental relationship between LCI and Curv due to CDI (Verbanck et al., 2012a). In the present work, we apply the same conceptual 2-compartment model to generate normalized phase III slopes (Sn) compatible with those measured experimentally, to suggest a method to deal with the abovementioned limitation of Scond (and Sacin) derived from such Sn curves.

We hypothesize two possible mechanisms of combined DCDI and CDI that could lead to Sn curves in cases of severe ventilation heterogeneity, including extreme cases leading to horizontal Sn asymptotes. These 2 hypothetical mechanisms and resulting relationships between Scond*, Sacin*, Curv and LCI, are then set against experimental data, to identify the most likely mechanism accounting for Sacin* and Scond* obtained in advanced CF lung disease. Hence the two aims of this study can be summarized as follows: (1) to develop an alternate method for Sacin and Scond calculation in cases of severe ventilation heterogeneity (leading to alternate parameters Sacin* and Scond*) and (2) to utilize a conceptual model to identify the various mechanisms of ventilation heterogeneity that can be captured by the alternate parameters Sacin* and Scond*.

Section snippets

Experiments

The study protocol was approved by the local research ethics committee (BUN 143201213405). We studied the MBW tests from 25 adult CF patients, for whom Curv and LCI were previously reported in relationship to their lung function status (Verbanck et al., 2012a). For the present study, an additional 20 CF patients were recruited in the same age range. As in our previous study (Verbanck et al., 2012a), standard lung function testing was performed (VMax encoreVE22d, Cardinalhealth®, Bilthoven, The

Experimental washout curves in CF patients

In Fig. 1, individual mean expired concentration and normalized slopes are shown for all 45 CF patients. These are set against the normal ranges (grey areas representing 95%CI) obtained from the 25 controls from our previous study (Verbanck et al., 2012a). For each patient, data are considered up to the point where mean expired concentration has reached the 1/40th threshold. The Sn curves also illustrate that the alternative computation of Scond* between TO = 0 and 3 (excluding the 1st breath),

Discussion

Ventilation heterogeneity can become severe in adult CF patients with Sn values of the last breath (just beyond the 1/40th threshold) that can be typically 5 times greater than in normal controls and sometimes even reach a horizontal asymptote (Fig. 1B). We propose that in cases of severe ventilation heterogeneity, the regression of Sn vs TO between TO = 1.5 and 6 (Scond) can be replaced by Scond*, which is the regression of Sn versus TO below TO = 3, but excluding the first breath to minimize

Conclusion

In summary, in cases of advanced CF lung disease, where LCI may be the result of irreversible functional changes across large portions of the lung, further refinement of the non-invasive diagnostic tools, such as the normalized phase III slope analysis, may be beneficial. We have proposed modified indices Scond* and Sacin* to enable a quantitative relationship to overall ventilation heterogeneity (LCI). While Scond* is probably related to the functional changes in large portions of the lung,

Acknowledgements

This study was funded by the Fund for Scientific Research – Flanders.

References (23)

  • A. Horsley

    Lung clearance index in the assessment of airways disease

    Respiratory Medicine

    (2009)
  • A.R. Horsley et al.

    Effects of cystic fibrosis lung disease on gas mixing indices derived from alveolar slope analysis

    Respiratory Physiology and Neurobiology

    (2008)
  • J.S. Brown et al.

    Effect of ventilation distribution on aerosol bolus dispersion and recovery

    Journal of Applied Physiology

    (1998)
  • A.B. Crawford et al.

    Effect of breath holding on ventilation maldistribution during tidal breathing in normal subjects

    Journal of Applied Physiology

    (1986)
  • A.B. Crawford et al.

    Convection- and diffusion-dependent ventilation maldistribution in normal subjects

    Journal of Applied Physiology

    (1985)
  • L.A. Engel

    Intraregional gas mixing and distribution

  • P.M. Gustafsson

    Peripheral airway involvement in CF and asthma compared by inert gas washout

    Pediatric Pulmonology

    (2007)
  • C.J. McMahon et al.

    Hyperpolarized 3helium magnetic resonance ventilation imaging of the lung in cystic fibrosis: comparison with high resolution CT and spirometry

    European Radiology

    (2006)
  • M. Paiva

    Two new pulmonary functional indexes suggested by a simple mathematical model

    Respiration

    (1975)
  • M. Paiva et al.

    Theoretical studies of gas mixing and ventilation distribution in the lung

    Physiological Reviews

    (1987)
  • R. Pellegrino et al.

    Interpretative strategies for lung function tests. Series ATS/ERS Task Force: Standardisation of lung function testing

    European Respiratory Journal

    (2005)
  • Cited by (36)

    • An innovative lung model for multiple breath washout testing in health and disease

      2019, Clinical Biomechanics
      Citation Excerpt :

      The adaptation of the model to conditions of advanced VI increased measured estimates of VI as hypothesized. LCI, SIII of SF6 expiratory curve and Sacin increased as observed in patients with chronic obstructive lung disease (Horsley et al., 2008; Stanojevic et al., 2015; Verbanck et al., 2013; Yammine et al., 2013). We consider that ventilation inhomogeneity in this lung model is driven both by convection (due to pressure gradient) and by diffusion (due to concentration gradient) (Chang, 1988; Weibel et al., 2005).

    • Long-Term Consequences of Childhood Respiratory Disease

      2019, Kendig's Disorders of the Respiratory Tract in Children
    • A simulation study of diffusion-convection interaction and its effect on multiple breath washout indices

      2018, Respiratory Physiology and Neurobiology
      Citation Excerpt :

      The tight relationship between LCI and Sacin predicted by the simulations was verified experimentally in the smokers under study, with LCI and Sacin sharing 72% of the variance (Fig. 6). Similar associations have been previously observed in experimental studies in asthma (Farrow et al., 2017; Kjellberg et al., 2016; Verbanck et al., 2012) and cystic fibrosis (Verbanck et al., 2013). The very different kind of underlying structural changes in these various disease groups leading to Sacin change are in line with our simulated findings where very different physical effects are seen to modify Sacin.

    • The multiple-breath nitrogen washout test

      2018, Lung Function Testing in the 21st Century: Methodologies and Tools Bridging Engineering to Clinical Practice
    View all citing articles on Scopus
    View full text