Progressive Decline in FRC in Infants: Physiology or Technology?/From the Authors

To the Editor:

H

To test our hypothesis, we compared FRC results obtained using the Jaeger plethysmograph, a conventional plethysmograph (Department of Medical Engineering, Hammersmith Hospital, London, UK) (4), and a helium gas dilution system (EBS 2615, Equilibrated Bio Systems, Smithtown, NY). Measurements were performed in vitro using a lung model filled with copper wool to minimize the adiabatic effect and in vivo on 10 sedated infants, median age of 13 (range 6-14) mo. Seven infants were born prematurely (median gestational age 29 wk, range 24-31 wk), four had had bronchopulmonary dysplasia, and three were born at term (one had unilateral pulmonary aplasia and two had had surgical repair of congenital anomalies).

In vitro, the volumes measured by the Jaeger plethysmograph were lower than those measured by the Hammersmith plethysmograph (p < 0.001) and the helium gas dilution system (p < 0.001). The measured volumes by the Jaeger plethysmograph were between 87% and 89% of the actual volume, by the Hammersmith plethysmograph between 99% and 103% of the actual volume, and by the helium gas dilution system between 99% and 101% of the actual volume. In vivo, the median FRC measured using the Jaeger plethysmograph (19.6, range 13.9-36.2 ml/kg) was lower than that using the Hammersmith system (26.5, range 21.7-37.8 ml/kg) (p = 0.02) and the helium gas dilution system (21.9, range 15.2-30.6 ml/kg) (p = 0.09).

Our in vivo results using the Jaeger plethysmograph were similar to those obtained by Hulskamp and coworkers (1) and were on average 25-30% less than those expected from predicted values (2). The data we report suggest that the Jaeger plethysmograph underrecords both in vivo and in vitro, possibly due to a greater sensitivity to the adiabatic effect. These results emphasize the importance of comprehensively validating all new lung function measurement systems before introducing them into clinical or research practice.

Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

SIMON BROUGHTON

GERRARD F. RAFFERTY

ANTHONY D. MILNER

ANNE GREENOUGH

King's College London

London, United Kingdom

References

1. Hülskamp G, Hoo AF, Ljungberg H, Lum S, Pillow JJ, Slocks J. Progressive decline in plethysmographic lung volumes in infants: physiology or technology? Am J Respir Crit Care Med 2003;168:1003-1009.

2. Stocks J. Quanjer PH. Reference values for residual volume, functional residual capacity and total lung capacity. ATS Workshop on Lung Volume Measurements. Official Statement of The European Respiratory Society. Ear Respir J 1995;8:492-506.

3. Castile RG, Iram D, McCoy KS. Gas trapping in normal infants and in infants with cystic fibrosis. Pediatr Pulmonol 2004;37:461-469.

4. Thomas MR, Rafferty GF, Limb ES, Peacock JL, Calvert SA, Marlow N, Milner AD, Greenough A. Pulmonary function at follow-up of very preterm infants from the UK oscillation study. Am J Respir Crit Care Med 2004;169:868-872.

From the Authors:

We welcome initiatives to assess and validate infant lung function equipment and would willingly participate in any attempt to resolve queries or potential discrepancies. Due to lack of detail regarding the lung model and protocol used for in vitro and in vivo evaluation of the Jaeger MasterScreen BabyBody plethysmograph by Broughlon and colleagues, it is difficult to comment on the results presented. Their data are in disagreement with previously published multicenter in vitro validation (1) of the same equipment using a carefully characterized and validated lung function model (2). In that study, volumes between 75 and 300 ml were measured with an accuracy of ± 2.5% (adhering to ATS/ERS guidelines [3]), and low volumes (50 ml or less) were overestimated by 9%. Accuracy was maintained over frequencies between 20-120 strokes per minute; hence the possibility of errors due to adiabatic artifacts seems unlikely. Careful direct comparison of the two lung models and the precise protocols would be necessary to resolve these differences.

In vivo validation of infant lung function equipment is notoriously difficult, and we commend Broughton and coworkers for attempting to undertake such comparisons. While it would have been helpful to know within-subject rather than group mean results, the observed differences between FRC obtained by the two infant plethysmographs are exactly as we would have predicted (4), and do not in themselves help to identify what is the "correct" value. As healthy infants cannot be sedated purely for such comparisons or methodological studies, a true "gold standard" for FRC^sub pleth^ in infants is nonexistent. We therefore wish to underline the key message of our original paper that the FRC^sub pleth^ data presented from 32 healthy infants are robust for the technique and protocol as described, and need to be augmented with far more measurements, data that we are currently collecting.

In contrast to measurements in adults, in whom repeat measures using different types of equipment and software can be undertaken relatively easily, it is extraordinarily difficult to exclude potential sources of systematic bias from infant lung function equipment. Therefore, thorough in vitro assessments, development of reference data for the specific techniques and equipments, sufficient published methodological information to enable researchers to reproduce results in their own setting and measurements in appropriate prospectively recruited healthy control subjects are mandatory when studying infants. Indeed, in the Castile study (5) quoted by Broughton and coworkers, interpretation of results would have been severely biased without a prospective healthy control group, in whom the mean (SD) FRC^sub pleth^ was found to be 1.1 (1.9) Z-scores lower than that predicted (i.e.. ~ 90%) from the authors' own reference data published 4 yr earlier using similar equipment and protocol (6).

The letter from Broughton and coworkers endorses the key message of our paper (4) that any equipment for infant lung function testing requires comprehensive validation both in vitro and in vivo, and that reference data not only needs to be specific for the equipment used, but should be continuously updated and re-evaluated in the light of technical and methodological progress. We would welcome the opportunity to undertake further collaborative work in this field.

Conflict of Interest Statement: G.H. received a Jaeger MasterScreen BabyBody plethysmograph from VIASYS Healthcare and organized a scientific meeting financed by Abbott Virology. J.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

GEORG HÜLSKAMP

JANET STOCKS

University of Münster

Münster, Germany

and

Institute of Child Health, London, United Kingdom

References

1. Reinmann B, Stocks J, Frey U. Assessment of an infant whole-body plethysmograph using an infant lung function model. Eur Respir J 2001;17: 765-772.

2. Frey U, Reinmann B, Stocks J. The infant lung function model: a mechanical analogue to test infant lung function equipment. Eur Respir J 2001; 17:755-764.

3. Frey U, Stocks J, Coates A, Sly P, Bates J. Standards for infant respiratory function testing: Specifications for equipment used for infant pulmonary function testing. Eur Respir J 2000;16:731-740.

4. Hülskamp G, Hoo AF, Ljungberg H, Lum S, Pillow JJ, Stocks J. Progressive decline in plethysmographic lung volumes in infants: physiology or technology? Am J Respir Crit Care Med 2003;168:1003-1009.

5. Castile RG, Iram D, McCoy KS. Gas trapping in normal infants and in infants with cystic fibrosis. Pediatr Pulmonol 2004;37:461-469.

6. Castile R, Filbrun D, Flucke R, Franklin W, McCoy K. Adult-type pulmonary function tests in infants without respiratory disease. Pediatr Pulmonol 2000;30:215-227.