Evaluation of ventilation maldistribution as an early indicator of lung disease in children with cystic fibrosis

Study subjects

All paediatric CF patients aged 3–18 yrs attending the regional CF centre in Gothenburg for annual checkup over a 3‐yr period were asked to participate in the study. All were in a stable condition with no signs of infective exacerbation at the time of test. A total of 28 healthy subjects (no history of wheeze, chronic cough or sputum production, asthma or allergy) of similar age and sex distribution were recruited as controls. All participants were of Caucasian origin. The study was approved by the Regional Ethics Committee for Human Research and informed oral consent was obtained from the parents and also from the participants themselves when applicable.

Study design

All children performed a minimum of three FVC manoeuvres, and FEV₁, maximum expiratory flow at 50% of FVC (MEF₅₀), and MEF₂₅ were recorded. The residual volume (RV) and total lung capacity (TLC) were determined plethysmographically in subjects aged ≥6 yrs. All subjects then performed a minimum of three washin-washouts of a gas mixture containing 4% sulphur hexafluoride (SF₆) and 4% helium (He). Two indices of ventilation inhomogeneity: the mixing ratio (MR); and the lung clearance index (LCI), were calculated. In the primary analysis, subjects were classified according to whether they had normal or abnormal spirometry or RV/TLC ratio, and prevalence of ventilation inhomogeneity was determined for each group. Secondary analyses determined correlation between different measures of lung function and age. The relationship between lung function findings and chronic airway colonisation by typical CF bacterial pathogens, pancreatic function and genotype, were assessed in the CF group. Chronic colonisation was defined as sputum or naso-pharyngeal cultures found to be positive repeatedly over a 6‐month period.

Methods

Summary

In this cross-sectional study spirometry and RV/TLC findings were compared to indices of ventilation inhomogeneity in a cohort of 43 children with CF and in a matched cohort of 28 control children. All control children had normal FEV₁ and, by definition, normal gas mixing indices. Ventilation inhomogeneity was detected in one-half to two-thirds of CF subjects with normal spirometry or RV/TLC results, and in almost all CF children aged <10 yrs who had normal FEV₁. Abnormal gas mixing was found in virtually all CF patients with abnormal spirometry findings, abnormal RV/TLC ratios, or chronic bacterial airway colonisation, as well as in two-thirds of those not colonised. The findings indicate that CF lung disease may start early in childhood and long before it becomes evident with the methods currently used for monitoring of lung function.

Previous multiple-breath inert gas washout studies in cystic fibrosis subjects

MBW is not a new technique, and several previous studies, mostly performed during the 1980s and early 1990s, have investigated ventilation distribution in CF subjects using this or related methods. These studies have already shown that indices of ventilation inhomogeneity are raised in CF subjects compared with healthy subjects 19–25, and that ventilation inhomogeneity is correlated to airway resistance 21, FEV₁19, and Schwachman score 22 in CF subjects. Furthermore, there is evidence that ventilation inhomogeneity decreases during treatment of an acute exacerbation in CF subjects 24. Despite these encouraging early data the MBW technique has not been adopted in clinical practice, as spirometry has been considered to be quicker and easier to perform, and to provide adequate information for clinical care. The impetus for the current study is the recognition that, as CF care becomes more aggressive, the characteristics of the paediatric CF population are changing, and that spirometry may not be sensitive enough to detect the earliest changes of CF lung disease 8. It is notable that the CF populations reported in most previous studies of MBW had markedly abnormal spirometry results, with mean FEV₁ reported as 53% pred 19, 20, 53.1% pred 24, and 68.2% pred 25. In contrast, the CF population reported here had a mean FEV₁ of 89.1% pred.

Methodological aspects

Only 43 of 70 eligible patients were included in the study. The reason why some families declined to participate was therather long journey to the lung function laboratory. At the time of the study the MBW tests were only performed at the Central Hospital of Skövde, which is situated 160 km from the regional CF center in Gothenburg. The spirometry results from the study group of CF patients who participated in the study are slightly but not significantly lower than the results obtained from paediatric CF patients at the centre as a whole. A recent review of the patient records from the paediatric CF population at this centre showed that the mean (sd) FEV₁ was 96 (14) % pred and that 82% of the patients had a normal FEV₁ (unpublished data).

Sex distribution and mean age, height and weight were similar for CF and control groups. However, the majority of the controls were aged 7–15 yrs, which was a slightly narrower age distribution than in the CF cohort. MR and LCI results were very similar throughout the age span in the healthy controls, suggesting that the upper limits of normality calculated for these parameters are likely to be valid even at aged <7 and >15 yrs. Ljungberg et al. 18 supported the view that normal values for LCI and MR are very similar for children aged <7 yrs, as compared with older children. The lack of change of these indices with age in the healthy controls is encouraging.

None of the CF patients in this study were considered chronically colonised with Staphylococcus aureus at the time of the test. This probably reflects the frequent use of prophylactic long-term treatment with flucloxacillin at this centre 26.

In this study SF₆ was used as an inert marker gas and a mass spectrometer for measuring the gas concentrations. Previous studies have used the N₂ MBW technique, where resident N₂ is washed out by 100% O₂. The results obtained from SF₆ washout are not expected to differ from those obtained from N₂ washout, but the newer technique does overcome one technical problem. As a gas mixture containing a high O₂ concentration has a higher dynamic viscosity than air, the velocity at which the gas sample is transported through the mass spectrometer capillary will vary significantly with changing O₂ concentration during a N₂ washout. These variations introduce errors in the alignment between the gas sample and the flow signal, which are difficult to correct for when using a mass spectrometer.

A mass spectrometer was used because it enables the concentration of several additional gases, such as He and CO₂, to be recorded simultaneously. Further analyses of recordings including these gases can be performed in the future when suitable software and algorithms are available, offering the potential for more detailed information about lung function. Because SF₆ and He differ in terms of diffusivity, a comparison of their washouts is expected to give additional information. In particular, a comparison of the phase III slope calculated from the plot of expired volume against marker gas concentration may give information aboutthe site of peripheral airway pathology. However, the authors have noted that the phase III slope is affected by age,body size, and V_T. Further methodological work is necessary before data from this study can be interpreted in this way 11,12.

Multiple-breath inert gas washout and spirometry

In the present study the authors noticed large differences in LCI and MR between subjects with similar FEV₁% pred. The most important finding was that many subjects with normal spirometry results had abnormal MR and LCI and that almost all with abnormal spirometry had abnormal ventilation distribution (table 5⇑). The majority of CF children aged <10 yrs and those without evidence of chronic lower airway bacterial colonisation fell into the first category, with the majority of these children having normal FEV₁ and MEF₂₅ results but abnormal MR and LCI results. These findings indicate that MR and LCI are probably both sensitive and specific with respect to CF airway disease.

Older CF subjects, with lower FEV₁ and MEF₂₅ results, tended also to have higher MR and LCI results, but this pattern was not consistent (fig. 1c, 1d⇑, 2a, 2b⇑). No significant correlation between age and MR or LCI was found, although a significant relationship was found between % pred FEV₁ or MEF₂₅ results and MR results, not only among children with CF, but also in the control group. Some CF subjects with abnormal spirometry results were found to have widely differing ventilation distribution indices. For example, in three CF patients with FEV₁ between 55–60 % pred, the MR varied from almost normal (1.4) to 3.0. This observation could be explained by genetic and environmental factors, plus intra-subject variability within the FEV₁ data. Given that a starting FEV₁ of anywhere between 78–122% pred can be considered normal, then a child with an FEV₁ of 60% pred could have lost anything between 22–51% of their starting FEV₁. An alternative, physiological, explanation for the incomplete correlation between FEV₁ and MBW arises from the observation that MBW is a tidal breathing test, whilst FEV₁ is a FVC test. If the airways leading to particularly diseased regions of a subject's lungs are closed during quiet breathing, then MR and LCI will underestimate ventilation maldistribution. Further studies, comparing results from techniques such as the vital capacity single-breath washout test 27 with MBW indices, are required to address this question.

Significance of abnormal gas mixing

What is the significance of an abnormal MR or LCI in a child with CF? Impaired ventilation distribution can result from asymmetric narrowing of the airway lumen at branch points throughout the airway tree, which in turn may be caused by inflammation, scarring or obstruction by mucus, be secondary to changes in airway tone, or inhomogeneity may result from parenchymal changes in the subtended lung units. From a clinical perspective, the most important consideration is whether impaired gas mixing represents early changes of CF lung disease, which will subsequently progress, or whether impaired gas mixing is a fixed epiphenomenon, present in most children with CF, but unconnected to future prognosis. An indirect support for inflammation/infection being responsible for the abnormal MBW findings is the fact that patients with chronic colonisation had more pathological MR and LCI results than noncolonised patients. It is well known that children with CF become infected early in life, long before the consequences of that can be detected by conventional lung function tests 28, 29. The question whether early infection and inflammation are manifested as abnormal gas mixing will only be answered by a combination of longitudinal studies, examining progression of ventilation inhomogeneity and spirometry with age, and intervention studies, examining the effect of treatment upon ventilation inhomogeneity. In the meantime, the observation by the authors of the relationship between high MR and low FEV₁ is worth noting, but should not be over interpreted.

There are two further points that suggest that MBW measurements may be of clinical value in CF. First, a number of centres have now reported on the use of MBW in infants 30–32, and the data reported in the current study support previous suggestions that the technique can also be adapted for use in preschoolchildren 22, 23. The finding that the parameters used for reporting the MBW studies appear to give similar values in healthy children of different age adds to the potential value of this method, as interpretation of forced expiratory parameters can be confounded by the effects of growth and development 33.

Secondly, further adaptation of the technique may provide more detailed evidence as to the site of early CF lung disease, and to the nature of its progression. At present, the site at which ventilation inhomogeneity is being generated insubjects with CF can only be speculated. Spirometry parameters are known to be insensitive to peripheral airways disease and CF disease is known to involve small as well as large airways. The authors suggest that the abnormal ventilation distribution seen in many of the subjects in the current study is the result of changes in the small airways, since these changes were observed in so many subjects with normal spirometry. However, it is also possible that a raised MR can be generated by changes in the large airways, which are not yet detectable by spirometry. In the future, analysis of the progression of the phase III slope through a MBW, and comparison of slopes obtained from gases of differing diffusivities, may allow the site within the airway tree where inhomogeneity is generated to be identified. These more sophisticated interpretations of the MBW are based upon modeling and experimental studies performed in adults, during controlled breathing 11, 12, 34. Some further methodological work is necessary before these analyses can be applied to measurements obtained during uncontrolled breathing in children.

Conclusions

Abnormal ventilation distribution is found in the majority of children with cystic fibrosis, including young children with normal spirometry measurements. These findings suggest that destructive processes in the airways of children with cystic fibrosis may start early and before they are evident with the methods currently used when monitoring lung function.