- Email: [email protected]
Influence of preterm onset of inspiration on tidal breathing parameters in infants with and without CLD
Respiratory Physiology & Neurobiology 135 (2003) 39 /46 www.elsevier.com/locate/resphysiol
Influence of preterm onset of inspiration on tidal breathing parameters in infants with and without CLD Gerd Schmalisch a,*, Roland R. Wauer a, Bertram Foitzik a, Andreas Patzak b a
b
Clinic of Neonatology (Charite´), Humboldt-University of Berlin, Schumannstraße 20/21, D-10098 Berlin, Germany Johannes-Mu¨ller-Institute of Physiology, Humboldt-University of Berlin, Schumannstraße 20/21, D-10098 Berlin, Germany Accepted 4 February 2003
Abstract The preterm onset of inspiration (POI) is a well-known breathing strategy in newborns to increase their endexpiratory lung volume. The aim of this study was to investigate to which extent POI is related to tidal breathing (TB) parameters in healthy neonates (n /54) and infants with chronic lung diseases (CLD, n /45) with same postconceptional age. Using the deadspace free flow-through technique, 10 /60 consecutive breaths were evaluated during quiet sleep and POI was derived from the averaged flow /volume loop considering the end-expiratory flow level. Respiratory rate (RR), ventilation (/V˙/E) and peak flows were significantly higher in CLD infants compared with controls. The incidence of POI did not differ significantly between both patient groups. POI is strongly associated with TB parameters describing the shape of flow profiles or flow /volume loops. In contrast, TB parameters, which depend only on breathing depth and rate (e.g., RR, VT, V˙/E), were not significantly associated. The study shows that in infants TB parameters describing the flow profile may reflect differences in breathing strategy rather than impaired respiratory functions. # 2003 Elsevier Science B.V. All rights reserved. Keywords: Control of breathing; Preterm onset of breathing; Development; Newborn breathing; Preterm onset; Mammals; Humans; Pattern of breathing; Newborn
1. Introduction An infant’s breathing pattern, measured during tidal breathing (TB), contains physiological information about respiratory control and pulmonary mechanics. For a quantitative evaluation of the
* Corresponding author. Tel.: /49-30-450-516104; fax: / 49-30-450-516921. E-mail address: [email protected] (G. Schmalisch).
breathing pattern, TB is commonly measured at the airway opening using a pneumotachograph (PNT) connected to a face mask (Bates et al., 2000). Since this technique can be used relatively easy in sick neonates at the bedside, TB measurements are used increasingly for clinical (Carlsen et al., 1997; Schmalisch et al., 2001b; Williams et al., 2000) and research purposes (Stocks et al., 1994; Ueda et al., 1999; van der Ent et al., 1998). However, TB measurements in infancy are influenced by the peculiarities of the breathing
1569-9048/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S1569-9048(03)00029-6
40
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
strategies in early postnatal age. Adults and older infants breathe from an end-expiratory lung volume determined by the opposing recoil of the lungs and chest wall. Neonates have a highly compliant chest wall which may cause several problems during breathing, e.g., small end-expiratory lung volume, low oxygen stores, and a high risk for airway occlusion and atelectasis (Stark et al., 1987). Therefore, infants compensate for this mechanical disadvantage by actively maintaining lung volume above the resting volume. Kosch and Stark (1984) have shown in the past that a preterm onset of inspiration (POI) before the complete expiration provide a neonatal breathing strategy to dynamically increase the lung volume. Different techniques can be used to detect a POI. In newborns, the onset of inspiration is commonly measured by surface diaphragmatic electromyography (Fox et al., 1988), by intrathoracic-impedance measurements (Nikischin et al., 1996) or by pneumotachography. In a clinical study investigating synchronized mechanical ventilation, Hummler et al. (1996) have recently shown that a flow /volume signal is less prone to artifacts and chest wall distortions compared with other techniques. Furthermore, it is well known from the measurement of respiratory mechanics by occlusion tests (LeSouef et al., 1984) that a POI can be recognized from the TB flow /volume loop (TBFVL) by a sharp break of the end-expiratory flow as shown in Fig. 1. Modern algorithms for infant respiratory function testing enable us to derive a great variety of TB parameters (Schmalisch et al., 1996), but the relation of a POI with different parameters was not yet investigated in detail. We hypothesize that a POI is associated with changes in different TB parameters. Therefore, the aim of this study was to investigate the relation of a POI with the most commonly measured TB parameters in comparison to the differences in TB parameters between healthy neonates and infants with chronic lung diseases (CLD) with the same postconceptional age. At present, infants with CLD represent the most important group for respiratory function testing in newborns and in early infancy (Bancalari, 2001; Jobe and Ikegami, 2000).
2. Materials and methods 2.1. Subjects In a prospective clinical study over a 25-month period, TB measurements were performed in 54 healthy neonates and 45 infants with CLD defined by mechanical ventilation /48 h and oxygen dependence at day 28 of life (Bancalari, 2001). All measurements were performed in the respiratory function laboratory of the Clinic of Neonatology at Humboldt University (Charite´). Inclusion criteria for this study were the postconceptional age B/50 weeks, spontaneous breathing, quiet sleep according to Prechtl (1974) and written parental consent. Infants with major malformations, upper airway obstruction, acute respiratory infection, congenital heart disease and central nervous system malfunction were not eligible. All healthy neonates and 37 infants with CLD were born in our hospital, whereas eight CLD infants came from different hospitals. As shown in Table 1, there were significant differences between both patient groups in birth weight and gestational age. However, at the day of measurement, both groups were comparable for postconceptional age and body weight. This study was approved by the Ethical Committee of the Medical Faculty (Charite´) of the Humboldt University (protocol 54/92). Parents were given a full explanation of the tests and equipment used before their written consent was obtained. 2.2. Equipment for TB measurements TB was measured using the deadspace free flowthrough technique as described previously (Schmalisch et al., 2001a). Briefly, the face mask is continuously rinsed by a constant background flow higher than the infant’s peak tidal inspiratory flow. The flow in and out of a modified transparent face mask (Vital Signs, Inc., Totowa, USA) were measured by two screen PNTs (Baby PNT Jaeger, Wuerzburg, Germany) with a low flow resistance (RPNT /0.2 kPa L 1 sec, total resistance defined by the back pressure at 5 L min 1). The infant’s tidal flow was measured by the
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
41
Fig. 1. TBFVLs of consecutive breathing cycles (left) and the averaged loop (right) of a healthy newborn (top) and an infant with CLD (bottom). At the beginning of inspiration, the volume is always reset to zero. The level of the end-expiratory flow break-off (/V˙ b-o ) as a marker of a POI was determined manually from the averaged loop. A temporary end-expiratory flow increase (bottom) was neglected.
difference between the two flow signals. Both PNTs were calibrated simultaneously at the beginning of each measurement using a 100 ml calibration syringe (Hans Rudolph, Kansas City, USA).
The flow signal was filtered by an analogue Bessel filter of fourth-order with a cut-off frequency of 48 Hz to avoid aliasing, sampled with a 16 bit analogue/digital converter and recorded at 200 Hz.
Table 1 Patient characteristics (median and range are given in brackets)
Birth weight (g) Gestational age (weeks) Age (days) Postconceptional age (weeks) Body weight at the time of measurement (g)
Healthy neonates (n /54)
Infants with CLD (n /45)
P -value
3165 (1610 /4880) 38 (31 /41) 7 (3 /14) 39.3 (31.7 /42.7) 3100 (1590 /4710)
890 (450 /3860) 27 (24 /34) 89 (33 /162) 40.0 (34.6 /49.0) 2800 (1950 /5300)
B/0.0001 B/0.0001 B/0.0001 0.15 0.08
42
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
2.3. Measurements in infants Most infants were studied during natural, quiet sleep assessed by behavioral criteria (Prechtl, 1974), but 15 (15%) were sedated with chloral hydrate (50 mg kg 1) given orally 15/30 min before testing. Sedation was primarily necessary in older infants and for additional lung function tests. In a previous study (Schmalisch et al., 2001a), we did not find a statistically significant influence of such mild sedation on measured TB parameters. Parents were usually present during the respiratory function testing. Sleeping infants were measured in a supine position with the neck in a neutral position, supported by a head seal. After a period of accommodation (5 /20 min), TB was measured while the airtight seal of the mask on the infant’s face was checked by continuous leak monitoring (Foitzik et al., 1998). The end of the accommodation period is commonly characterized by a more regular breathing pattern without any visible drift in breathing parameters (Patzak et al., 2001). The graphical display of instantaneous respiratory rate (RR) over the last 60 breathing cycles also contributes to assessing of the stability of the TB parameters. The duration of the TB measurements was normally 20 /30 min depending on the period of accommodation to the face mask. Infants were continuously monitored by pulse oximetry. Depending on the variability of the breathing pattern, at least 10 but not more than 60 consecutive breaths with the same basic pattern of the TBFVLs were selected by the investigator and evaluated. The median of evaluated cycles in healthy and CLD infants was 18 and 26, respectively. This difference was not statistically significant. The averaged TBFVL of all selected breathing cycles was calculated to determine a characteristic loop pattern as shown previously (Schmalisch et al., 2001b). The shape of the expiratory limb after the peak flow was classified in monotone decrease (linear, concave and convex), flow braking with nearly constant expiratory flow or multimodal courses (expiratory limb with more than one flow peak excluding a small expiratory flow peak at the end of expiration as shown in Fig. 1). For the quantitative evaluation,
seven basic parameters were measured from the flow and volume signals (inspiratory time (tI), expiratory time (tE), tidal volume (VT), peak tidal expiratory flow (PTEF), time to PTEF (tPTEF), volume to PTEF (VPTEF) and flow when 25% of tidal volume remains in the lung (TEF25)). From these parameters, seven characteristic TB parameters were derived (RR, minute ventilation (/V˙/E), mean inspiratory flow VT/tI, mean initial expiratory gas acceleration PTEF/tPTEF, and the ratios tPTEF/tE, VPTEF/VT and TEF25/PTEF). The level of the end-expiratory flow break-off (/V˙ b-o ) derived from the averaged TBFVL was taken to recognize a POI. As shown in Fig. 1, V˙ b-o was defined by the point of maximal change in the descent of the TBFVL in the transition from expiration to inspiration. V˙ b-o was determined manually by one experienced investigator. Only in few cases were two experts consulted for those loops where V˙ b-o could not be clearly identified. V˙ b-o was classified in three groups: V˙ b-o lower than 1/3 PTEF, between 1/3 and 2/3 PTEF or greater than 2/3 PTEF. In few infants at the end of expiration, a small flow peak occurred (Fig. 1). In these cases, the flow before this end-expiratory flow peak was taken (Fig. 1). A distinct POI was defined by V˙ b-o/ /1/3 PTEF. 2.4. Statistical methods Patient characteristics are recorded as the median and range, and compared using the Wilcoxon and Mann /Whitney tests. Differences in the pattern of the TBFVL between the groups were tested by means of the x2-test. Mean and standard deviations (S.D.) were calculated for all TB parameters in both patient groups. A multifactor analysis of variance (MANOVA) was used to test the statistical significance of both factors CLD and POI on TB parameters. A level of statistical significance of P B/0.05 was accepted.
3. Results Between both patient groups, there were significant differences (P B/0.01) in the shape of the expiratory limb of the TBFVL (Fig. 2). In CLD
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
43
Fig. 2. Distribution of typical shapes of the expiratory limb of the TBFVL after PTEF in both patient groups.
infants, the incidence of concave shapes was 27% and nearly 10 times higher than in healthy neonates (3%). The comparison of TB parameters measured in both patient groups is shown in Table 2. There were statistically significant differences between both patient groups except for tidal volume and endexpiratory flow (TEF25) related to body weight. The largest difference was in RR, minute ventilation V˙/E and the mean inspiratory flow VT/tI.
The level of V˙ b-o related to PTEF showed the same distribution in both patient groups (Fig. 3). V˙ b-o was determined from the averaged TBFVL and there was no information about the breath-tobreath variability. However, the within-subject variability of flow at the end of expiration was measured by TEF25. The median (range) coefficient of variation of TEF25 was 16% (8 /26%) in healthy infants and 15% (4 /24%) in CLD infants without statistically significant difference.
Table 2 Comparison of TB parameters measured in healthy neonates and infants with CLD (presented are group means9/S.D.) Parameter
Healthy neonates (n /54)
CLD infants (n /45)
P -value for CLD
P -value for POI
RR (min 1) VT (ml kg 1) ˙/E (ml min 1 kg 1) /V PTEF (L min 1 kg 1) VT/tI (ml sec 1 kg 1) tPTEF (msec) tPTEF/tE (%) VPTEF/VT (%) PTEF/tPTEF (L sec 2 kg 1) TEF25 (L min 1 kg 1) TEF25/PTEF
40.29/9.5 5.589/1.0 2219/58.0 0.669/0.21 9.139/2.4 2779/141 26.69/10.7 29.89/7.4 3.169/3.13 0.399/0.12 0.599/0.12
56.69/14.8 5.339/1.6 2939/76.4 0.919/0.29 11.139/2.7 1589/86 20.99/7.8 26.99/6.5 7.059/4.02 0.489/0.17 0.539/0.13
B/0.0001 0.85 B/0.0001 0.001 B/0.0001 B/0.0001 0.004 0.008 0.002 0.35 0.04
0.37 0.58 0.14 0.02 0.11 B/0.0001 B/0.0001 B/0.0001 B/0.0001 0.01 B/0.0001
Abbreviations of TB parameters: RR, respiratory rate; VT, tidal volume; V˙/E, minute ventilation; PTEF, peak tidal expiratory flow; VT/tI, mean inspiratory flow; PTEF/tPTEF, initial expiratory gas acceleration; tPTEF, time to PTEF; tPTEF/tE, time to PTEF as a proportion of tidal expiratory time; VPTEF/VT, volume to PTEF as a portion of tidal volume; TEF25, flow when 25% of tidal volume remains in the lung.
44
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
ments using the parameter tPTEF/tE. Obviously, in CLD infants, the parameter tPTEF/tE was lower than in healthy infants if V˙ b-o was the same. However, changes in V˙ b-o had a stronger influence on tPTEF/tE than the impaired respiratory functions of the CLD infants.
4. Discussion
Fig. 3. Distribution of the level of the end-expiratory flow break-off (/V˙ b-o ) in both patient groups.
A distinct POI (/V˙ b-o/ /1/3 PTEF) was seen in 25 (46%) of healthy newborns and in 14 (31%) of CLD infants. The difference was not statistically significant. As shown in Table 2, there is a significant relationship between POI, described by the three levels of V˙ b-o ; and many TB parameters especially those which describe the shape of the expiratory limb of the TBFVL, e.g., the ascent to PTEF (PTEF/tPTEF), the site of the PTEF (tPTEF/tE, VPTEF/VT) or the end-expiratory flow (TEF25/PTEF). The TB parameters depending only on breathing depth and rate, namely RR, VT, V˙/E and VT/tI were not affected significantly by POI. Fig. 4 shows an example of the clinical importance of considering POI in TB measure-
Fig. 4. Influence of V˙ b-o on the TB parameter tPTEF/tE in healthy infants and infants with CLD (**P B/0.01).
The main goal of this study was to investigate the relation of a POI with different TB measurements. We found that a POI is related mainly to those TB parameters describing the flow profile or the shape of the TBFVL (e.g., the extensively investigated parameter tPTEF/tE). In contrast, conventional TB parameters, which describe only rate and depth of breathing (RR, VT and V˙/E), are relatively independent on POI. The mechanisms explaining the association between POI and TB pattern in newborns are not fully understood. It is not easy to distinguish cause from effect. Most scientists would argue that it is not the POI that causes changes in breathing patterns described by TB parameters but vice versa. Therefore, a strong relationship between the two becomes almost inevitable. Indeed, the braking of expiratory flow together with a shortening of the expiratory time is well recognized as the means by which neonates defend lung volume (Stark et al., 1987). For the clinical use of TB measurements, it is important to know, however, that the POI is associated with significant changes in some TB parameters. In a neonatal lung function laboratory, infants with CLD constitute one of the most important patient groups. Up to now, CLD is the most common respiratory complication in preterm infants requiring prolonged mechanical ventilation (Jobe and Ikegami, 1998). In contrast to expectations, the improved mechanical ventilation and the introduction of surfactant for the treatment of an immature surfactant-deficient lung in preterm infants did not decrease the incidence of CLD, both factors may even have augmented CLD incidence due to the markedly increased survival of the smallest infants (Bancalari and del Moral, 2001).
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
In this study, the differences in TB parameters between both patient groups are in good agreement with published results. Tepper et al. (1986) reported a significantly higher RR in CLD infants compared with controls, but no significant changes in VT. In this study, the unchanged tidal volume related to body weight and the much higher RR in CLD infants compared to healthy neonates explain the significantly higher parameters describing expiratory flow pattern in infants with CLD. In contrast to TB, it is well recognized that at forced expiration, CLD infants show a significant flow limitation due to a poor growth of the airways and the resulting higher peripheral airway resistance (Tepper et al., 1986). In CLD infants of the present study, end-expiratory flow (TEF25) was not reduced. This was probably due to their high RR and the resulting higher expiratory flow rates. The parameters tPTEF, tPTEF/tE as well as VPTEF/VT, which is strongly correlated with tPTEF/tE describe the site of PTEF in the flow signal as well as in the TBFVL. These parameters are frequently used to detect airway obstructions (Martinez et al., 1988; Greenough et al., 1998; Lodrup et al., 1999). However, the association of these parameters with small airway caliber remains speculative and it could not be demonstrated in previously published studies (Aston et al., 1994; Williams et al., 2000). van der Ent et al. (1998) have shown that tPTEF/tE is much more influenced by post-inspiratory muscle activity than by lung mechanics. In this study, we found a slight but statistically significant reduction of tPTEF/tE in CLD infants. The higher PTEF and the shorter tPTEF in CLD infants explain the large differences between both patient groups in the mean initial gas acceleration given by PTEF/tPTEF. An unexpected finding of our study was that conventionally used TB parameters (e.g., RR, VT, V˙/E, mean/V˙/I), which are clearly defined and easy to derive, are relatively independent on POI. Therefore, in contrast to investigations in the past, where tPTEF/tE was the centre of interest (Carlsen et al., 1997; Greenough et al., 1998; Lodrup et al., 1999), TB measurements in neonates should be focused on the evaluation of these conventional
45
parameters measured under standardized conditions (Bates et al., 2000). A distinct POI with a V˙ b-o/ /1/3 PTEF was found in 41% of all infants. This means that a POI is a frequent breathing strategy in healthy and sick infants. Because of this high incidence and its significant influence on some TB parameters, it has to be taken into consideration when analyzing and interpreting TB parameters in newborns. The determination of a POI from TBFVL is subjected to methodological limitations. We have not tested by electromyography whether an increased V˙ b-o indicates a POI in all cases. It is well known from commonly used occlusion tests to measure respiratory mechanics (LeSouef et al., 1984; Gappa et al., 2001) that a V˙ b-o is a good indicator for an incomplete exhalation. In infants with flow braking, an increased end-expiratory flow must not be caused by POI. As shown in Fig. 2, such a pattern was seen in healthy and CLD infants in only 15 and 9%, respectively. In conclusion, a POI affects the interpretation of some TB parameters measured in newborns. The study shows that at this age, TB parameters describing the flow profile may reflect differences in the breathing strategy rather than impaired respiratory functions.
Acknowledgements The authors thank Dr. Mario Schmidt for his support in the development of the software and Mrs. Silke Schmidt for her assistance in respiratory function testing. They also gratefully acknowledge Prof. Colin Morley (Melbourne) for the critical review of the manuscript. This work was supported by the German Ministry for Education and Research, project ‘‘Perinatal Lung’’ (grant 01-ZZ-9511).
References Aston, H., Clarke, J., Silverman, M., 1994. Are tidal breathing indices useful in infant bronchial challenge tests? Pediatr. Pulmonol. 17, 225 /230.
46
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
Bancalari, E., 2001. Changes in the pathogenesis and prevention of chronic lung disease of prematurity. Am. J. Perinatol. 18, 1 /9. Bancalari, E., del Moral, T., 2001. Bronchopulmonary dysplasia and surfactant. Biol. Neonate. 80 (Suppl. S1), 7 /13. Bates, J.H., Schmalisch, G., Filbrun, D., Stocks, J., 2000. Tidal breath analysis for infant pulmonary function testing: ERS/ ATS task force on standards for infant respiratory function testing. Eur. Respir. J. 16, 1180 /1192. Carlsen, K.C.L., Stenzler, A., Carlsen, K.H., 1997. Determinants of tidal flow volume loop indices in neonates and children with and without asthma. Pediatr. Pulmonol. 24, 391 /396. Foitzik, B., Schmidt, M., Windstetter, D., Wauer, R.R., Schmalisch, G., 1998. Leak measurements in spontaneously breathing premature newborns by using the flow-through technique. J. Appl. Physiol. 85, 1187 /1193. Fox, R.E., Kosch, P.C., Feldman, H.A., Stark, A.R., 1988. Control of inspiratory duration in premature infants. J. Appl. Physiol. 64, 2597 /2604. Gappa, M., Colin, A.A., Goetz, I., Stocks, J., 2001. Passive respiratory mechanics: the occlusion techniques. Eur. Respir. J. 17, 141 /148. Greenough, A., Zhang, Y.X., Yuksel, B., Dimitriou, G., 1998. Assessment of prematurely born children at follow-up using a tidal breathing parameter. Physiol. Meas. 19, 111 /116. Hummler, H.D., Gerhardt, T., Gonzalez, A., Bolivar, J., Claure, N., Everett, R., Bancalari, E., 1996. Patienttriggered ventilation in neonates: comparison of a flowand an impedance-triggered system. Am. J. Respir. Crit. Care Med. 154, 1049 /1054. Jobe, A.H., Ikegami, M., 1998. Mechanisms initiating lung injury in the preterm. Early Hum. Dev. 53 (1), 81 /94. Jobe, A.H., Ikegami, M., 2000. Lung development and function in preterm infants in the surfactant treatment era. Annu. Rev. Physiol. 62, 825 /846. Kosch, P.C., Stark, A.R., 1984. Dynamic maintenance of endexpiratory lung volume in full-term infants. J. Appl. Physiol. 57, 1126 /1133. LeSouef, P.N., England, S.J., Bryan, A.C., 1984. Passive respiratory mechanics in newborns and children. Am. Rev. Respir. Dis. 129, 552 /556. Lodrup, C.K., Carlsen, K.H., Nafstad, P., Bakketeig, L., 1999. Perinatal risk factors for recurrent wheeze in early life. Pediatr. Allergy Immunol. 10, 89 /95.
Martinez, F.D., Morgan, W.J., Wright, A.L., Holberg, C.J., Taussig, L.M., 1988. Diminished lung function as a predisposing factor for wheezing respiratory illness in infants. N. Engl. J. Med. 319, 1112 /1117. Nikischin, W., Gerhardt, T., Everett, R., Gonzalez, A., Hummler, H., Bancalari, E., 1996. Patient-triggered ventilation: a comparison of tidal volume and chest wall and abdominal motion as trigger signals. Pediatr. Pulmonol. 22, 28 /34. Patzak, A., Foitzik, B., Mrowka, R., Schmalisch, G., 2001. Time of measurement influences the variability of tidal breathing parameters in healthy and sick infants. Respir. Physiol. 128, 187 /194. Prechtl, H.F., 1974. The behavioural states of the newborn infant (a review). Brain Res. 76, 185 /212. Schmalisch, G., Grubba, K., Wauer, R.R., 1996. Computeraided interpretation of pulmonary parameters in newborn infants. Med. Biol. Eng. Comput. 34, 160 /164. Schmalisch, G., Foitzik, B., Wauer, R.R., Stocks, J., 2001a. Effect of apparatus deadspace on breathing parameters in newborns: ‘‘flow-through’’ versus conventional techniques. Eur. Respir. J. 17, 108 /114. Schmalisch, G., Schmidt, M., Foitzik, B., 2001b. Novel technique to average breathing loops for infant respiratory function testing. Med. Biol. Eng. Comput. 39, 688 /693. Stark, A.R., Cohlan, B.A., Waggener, T.B., Frantz, I.D., Kosch, P.C., 1987. Regulation of end-expiratory lung volume during sleep in premature infants. J. Appl. Physiol. 62, 1117 /1123. Stocks, J., Dezateux, C.A., Jackson, E.A., Hoo, A.F., Costeloe, K.L., Wade, A.M., 1994. Analysis of tidal breathing parameters in infancy: how variable is tPTEF/tE. Am. J. Respir. Crit. Care Med. 150, 1347 /1354. Tepper, R.S., Morgan, W.J., Cota, K., Taussig, L.M., 1986. Expiratory flow limitation in infants with bronchopulmonary dysplasia. J. Pediatr. 109, 1040 /1046. Ueda, Y., Stick, S.M., Hall, G., Sly, P.D., 1999. Control of breathing in infants born to smoking mothers. J. Pediatr. 135, 226 /232. van der Ent, C.K., van der Grinten, C.P.M., Meessen, N.E.L., Luijendijk, S.C.M., Mulder, P.G.H., Bogaard, J.M., 1998. Time to peak tidal expiratory flow and the neuromuscular control of expiration. Eur. Respir. J. 12, 646 /652. Williams, E.M., Madgwick, R.G., Thomson, A.H., Morris, M.J., 2000. Expiratory airflow patterns in children and adults with cystic fibrosis. Chest 117, 1078 /1084.
Influence of preterm onset of inspiration on tidal breathing parameters in infants with and without CLD Gerd Schmalisch a,*, Roland R. Wauer a, Bertram Foitzik a, Andreas Patzak b a
b
Clinic of Neonatology (Charite´), Humboldt-University of Berlin, Schumannstraße 20/21, D-10098 Berlin, Germany Johannes-Mu¨ller-Institute of Physiology, Humboldt-University of Berlin, Schumannstraße 20/21, D-10098 Berlin, Germany Accepted 4 February 2003
Abstract The preterm onset of inspiration (POI) is a well-known breathing strategy in newborns to increase their endexpiratory lung volume. The aim of this study was to investigate to which extent POI is related to tidal breathing (TB) parameters in healthy neonates (n /54) and infants with chronic lung diseases (CLD, n /45) with same postconceptional age. Using the deadspace free flow-through technique, 10 /60 consecutive breaths were evaluated during quiet sleep and POI was derived from the averaged flow /volume loop considering the end-expiratory flow level. Respiratory rate (RR), ventilation (/V˙/E) and peak flows were significantly higher in CLD infants compared with controls. The incidence of POI did not differ significantly between both patient groups. POI is strongly associated with TB parameters describing the shape of flow profiles or flow /volume loops. In contrast, TB parameters, which depend only on breathing depth and rate (e.g., RR, VT, V˙/E), were not significantly associated. The study shows that in infants TB parameters describing the flow profile may reflect differences in breathing strategy rather than impaired respiratory functions. # 2003 Elsevier Science B.V. All rights reserved. Keywords: Control of breathing; Preterm onset of breathing; Development; Newborn breathing; Preterm onset; Mammals; Humans; Pattern of breathing; Newborn
1. Introduction An infant’s breathing pattern, measured during tidal breathing (TB), contains physiological information about respiratory control and pulmonary mechanics. For a quantitative evaluation of the
* Corresponding author. Tel.: /49-30-450-516104; fax: / 49-30-450-516921. E-mail address: [email protected] (G. Schmalisch).
breathing pattern, TB is commonly measured at the airway opening using a pneumotachograph (PNT) connected to a face mask (Bates et al., 2000). Since this technique can be used relatively easy in sick neonates at the bedside, TB measurements are used increasingly for clinical (Carlsen et al., 1997; Schmalisch et al., 2001b; Williams et al., 2000) and research purposes (Stocks et al., 1994; Ueda et al., 1999; van der Ent et al., 1998). However, TB measurements in infancy are influenced by the peculiarities of the breathing
1569-9048/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S1569-9048(03)00029-6
40
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
strategies in early postnatal age. Adults and older infants breathe from an end-expiratory lung volume determined by the opposing recoil of the lungs and chest wall. Neonates have a highly compliant chest wall which may cause several problems during breathing, e.g., small end-expiratory lung volume, low oxygen stores, and a high risk for airway occlusion and atelectasis (Stark et al., 1987). Therefore, infants compensate for this mechanical disadvantage by actively maintaining lung volume above the resting volume. Kosch and Stark (1984) have shown in the past that a preterm onset of inspiration (POI) before the complete expiration provide a neonatal breathing strategy to dynamically increase the lung volume. Different techniques can be used to detect a POI. In newborns, the onset of inspiration is commonly measured by surface diaphragmatic electromyography (Fox et al., 1988), by intrathoracic-impedance measurements (Nikischin et al., 1996) or by pneumotachography. In a clinical study investigating synchronized mechanical ventilation, Hummler et al. (1996) have recently shown that a flow /volume signal is less prone to artifacts and chest wall distortions compared with other techniques. Furthermore, it is well known from the measurement of respiratory mechanics by occlusion tests (LeSouef et al., 1984) that a POI can be recognized from the TB flow /volume loop (TBFVL) by a sharp break of the end-expiratory flow as shown in Fig. 1. Modern algorithms for infant respiratory function testing enable us to derive a great variety of TB parameters (Schmalisch et al., 1996), but the relation of a POI with different parameters was not yet investigated in detail. We hypothesize that a POI is associated with changes in different TB parameters. Therefore, the aim of this study was to investigate the relation of a POI with the most commonly measured TB parameters in comparison to the differences in TB parameters between healthy neonates and infants with chronic lung diseases (CLD) with the same postconceptional age. At present, infants with CLD represent the most important group for respiratory function testing in newborns and in early infancy (Bancalari, 2001; Jobe and Ikegami, 2000).
2. Materials and methods 2.1. Subjects In a prospective clinical study over a 25-month period, TB measurements were performed in 54 healthy neonates and 45 infants with CLD defined by mechanical ventilation /48 h and oxygen dependence at day 28 of life (Bancalari, 2001). All measurements were performed in the respiratory function laboratory of the Clinic of Neonatology at Humboldt University (Charite´). Inclusion criteria for this study were the postconceptional age B/50 weeks, spontaneous breathing, quiet sleep according to Prechtl (1974) and written parental consent. Infants with major malformations, upper airway obstruction, acute respiratory infection, congenital heart disease and central nervous system malfunction were not eligible. All healthy neonates and 37 infants with CLD were born in our hospital, whereas eight CLD infants came from different hospitals. As shown in Table 1, there were significant differences between both patient groups in birth weight and gestational age. However, at the day of measurement, both groups were comparable for postconceptional age and body weight. This study was approved by the Ethical Committee of the Medical Faculty (Charite´) of the Humboldt University (protocol 54/92). Parents were given a full explanation of the tests and equipment used before their written consent was obtained. 2.2. Equipment for TB measurements TB was measured using the deadspace free flowthrough technique as described previously (Schmalisch et al., 2001a). Briefly, the face mask is continuously rinsed by a constant background flow higher than the infant’s peak tidal inspiratory flow. The flow in and out of a modified transparent face mask (Vital Signs, Inc., Totowa, USA) were measured by two screen PNTs (Baby PNT Jaeger, Wuerzburg, Germany) with a low flow resistance (RPNT /0.2 kPa L 1 sec, total resistance defined by the back pressure at 5 L min 1). The infant’s tidal flow was measured by the
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
41
Fig. 1. TBFVLs of consecutive breathing cycles (left) and the averaged loop (right) of a healthy newborn (top) and an infant with CLD (bottom). At the beginning of inspiration, the volume is always reset to zero. The level of the end-expiratory flow break-off (/V˙ b-o ) as a marker of a POI was determined manually from the averaged loop. A temporary end-expiratory flow increase (bottom) was neglected.
difference between the two flow signals. Both PNTs were calibrated simultaneously at the beginning of each measurement using a 100 ml calibration syringe (Hans Rudolph, Kansas City, USA).
The flow signal was filtered by an analogue Bessel filter of fourth-order with a cut-off frequency of 48 Hz to avoid aliasing, sampled with a 16 bit analogue/digital converter and recorded at 200 Hz.
Table 1 Patient characteristics (median and range are given in brackets)
Birth weight (g) Gestational age (weeks) Age (days) Postconceptional age (weeks) Body weight at the time of measurement (g)
Healthy neonates (n /54)
Infants with CLD (n /45)
P -value
3165 (1610 /4880) 38 (31 /41) 7 (3 /14) 39.3 (31.7 /42.7) 3100 (1590 /4710)
890 (450 /3860) 27 (24 /34) 89 (33 /162) 40.0 (34.6 /49.0) 2800 (1950 /5300)
B/0.0001 B/0.0001 B/0.0001 0.15 0.08
42
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
2.3. Measurements in infants Most infants were studied during natural, quiet sleep assessed by behavioral criteria (Prechtl, 1974), but 15 (15%) were sedated with chloral hydrate (50 mg kg 1) given orally 15/30 min before testing. Sedation was primarily necessary in older infants and for additional lung function tests. In a previous study (Schmalisch et al., 2001a), we did not find a statistically significant influence of such mild sedation on measured TB parameters. Parents were usually present during the respiratory function testing. Sleeping infants were measured in a supine position with the neck in a neutral position, supported by a head seal. After a period of accommodation (5 /20 min), TB was measured while the airtight seal of the mask on the infant’s face was checked by continuous leak monitoring (Foitzik et al., 1998). The end of the accommodation period is commonly characterized by a more regular breathing pattern without any visible drift in breathing parameters (Patzak et al., 2001). The graphical display of instantaneous respiratory rate (RR) over the last 60 breathing cycles also contributes to assessing of the stability of the TB parameters. The duration of the TB measurements was normally 20 /30 min depending on the period of accommodation to the face mask. Infants were continuously monitored by pulse oximetry. Depending on the variability of the breathing pattern, at least 10 but not more than 60 consecutive breaths with the same basic pattern of the TBFVLs were selected by the investigator and evaluated. The median of evaluated cycles in healthy and CLD infants was 18 and 26, respectively. This difference was not statistically significant. The averaged TBFVL of all selected breathing cycles was calculated to determine a characteristic loop pattern as shown previously (Schmalisch et al., 2001b). The shape of the expiratory limb after the peak flow was classified in monotone decrease (linear, concave and convex), flow braking with nearly constant expiratory flow or multimodal courses (expiratory limb with more than one flow peak excluding a small expiratory flow peak at the end of expiration as shown in Fig. 1). For the quantitative evaluation,
seven basic parameters were measured from the flow and volume signals (inspiratory time (tI), expiratory time (tE), tidal volume (VT), peak tidal expiratory flow (PTEF), time to PTEF (tPTEF), volume to PTEF (VPTEF) and flow when 25% of tidal volume remains in the lung (TEF25)). From these parameters, seven characteristic TB parameters were derived (RR, minute ventilation (/V˙/E), mean inspiratory flow VT/tI, mean initial expiratory gas acceleration PTEF/tPTEF, and the ratios tPTEF/tE, VPTEF/VT and TEF25/PTEF). The level of the end-expiratory flow break-off (/V˙ b-o ) derived from the averaged TBFVL was taken to recognize a POI. As shown in Fig. 1, V˙ b-o was defined by the point of maximal change in the descent of the TBFVL in the transition from expiration to inspiration. V˙ b-o was determined manually by one experienced investigator. Only in few cases were two experts consulted for those loops where V˙ b-o could not be clearly identified. V˙ b-o was classified in three groups: V˙ b-o lower than 1/3 PTEF, between 1/3 and 2/3 PTEF or greater than 2/3 PTEF. In few infants at the end of expiration, a small flow peak occurred (Fig. 1). In these cases, the flow before this end-expiratory flow peak was taken (Fig. 1). A distinct POI was defined by V˙ b-o/ /1/3 PTEF. 2.4. Statistical methods Patient characteristics are recorded as the median and range, and compared using the Wilcoxon and Mann /Whitney tests. Differences in the pattern of the TBFVL between the groups were tested by means of the x2-test. Mean and standard deviations (S.D.) were calculated for all TB parameters in both patient groups. A multifactor analysis of variance (MANOVA) was used to test the statistical significance of both factors CLD and POI on TB parameters. A level of statistical significance of P B/0.05 was accepted.
3. Results Between both patient groups, there were significant differences (P B/0.01) in the shape of the expiratory limb of the TBFVL (Fig. 2). In CLD
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
43
Fig. 2. Distribution of typical shapes of the expiratory limb of the TBFVL after PTEF in both patient groups.
infants, the incidence of concave shapes was 27% and nearly 10 times higher than in healthy neonates (3%). The comparison of TB parameters measured in both patient groups is shown in Table 2. There were statistically significant differences between both patient groups except for tidal volume and endexpiratory flow (TEF25) related to body weight. The largest difference was in RR, minute ventilation V˙/E and the mean inspiratory flow VT/tI.
The level of V˙ b-o related to PTEF showed the same distribution in both patient groups (Fig. 3). V˙ b-o was determined from the averaged TBFVL and there was no information about the breath-tobreath variability. However, the within-subject variability of flow at the end of expiration was measured by TEF25. The median (range) coefficient of variation of TEF25 was 16% (8 /26%) in healthy infants and 15% (4 /24%) in CLD infants without statistically significant difference.
Table 2 Comparison of TB parameters measured in healthy neonates and infants with CLD (presented are group means9/S.D.) Parameter
Healthy neonates (n /54)
CLD infants (n /45)
P -value for CLD
P -value for POI
RR (min 1) VT (ml kg 1) ˙/E (ml min 1 kg 1) /V PTEF (L min 1 kg 1) VT/tI (ml sec 1 kg 1) tPTEF (msec) tPTEF/tE (%) VPTEF/VT (%) PTEF/tPTEF (L sec 2 kg 1) TEF25 (L min 1 kg 1) TEF25/PTEF
40.29/9.5 5.589/1.0 2219/58.0 0.669/0.21 9.139/2.4 2779/141 26.69/10.7 29.89/7.4 3.169/3.13 0.399/0.12 0.599/0.12
56.69/14.8 5.339/1.6 2939/76.4 0.919/0.29 11.139/2.7 1589/86 20.99/7.8 26.99/6.5 7.059/4.02 0.489/0.17 0.539/0.13
B/0.0001 0.85 B/0.0001 0.001 B/0.0001 B/0.0001 0.004 0.008 0.002 0.35 0.04
0.37 0.58 0.14 0.02 0.11 B/0.0001 B/0.0001 B/0.0001 B/0.0001 0.01 B/0.0001
Abbreviations of TB parameters: RR, respiratory rate; VT, tidal volume; V˙/E, minute ventilation; PTEF, peak tidal expiratory flow; VT/tI, mean inspiratory flow; PTEF/tPTEF, initial expiratory gas acceleration; tPTEF, time to PTEF; tPTEF/tE, time to PTEF as a proportion of tidal expiratory time; VPTEF/VT, volume to PTEF as a portion of tidal volume; TEF25, flow when 25% of tidal volume remains in the lung.
44
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
ments using the parameter tPTEF/tE. Obviously, in CLD infants, the parameter tPTEF/tE was lower than in healthy infants if V˙ b-o was the same. However, changes in V˙ b-o had a stronger influence on tPTEF/tE than the impaired respiratory functions of the CLD infants.
4. Discussion
Fig. 3. Distribution of the level of the end-expiratory flow break-off (/V˙ b-o ) in both patient groups.
A distinct POI (/V˙ b-o/ /1/3 PTEF) was seen in 25 (46%) of healthy newborns and in 14 (31%) of CLD infants. The difference was not statistically significant. As shown in Table 2, there is a significant relationship between POI, described by the three levels of V˙ b-o ; and many TB parameters especially those which describe the shape of the expiratory limb of the TBFVL, e.g., the ascent to PTEF (PTEF/tPTEF), the site of the PTEF (tPTEF/tE, VPTEF/VT) or the end-expiratory flow (TEF25/PTEF). The TB parameters depending only on breathing depth and rate, namely RR, VT, V˙/E and VT/tI were not affected significantly by POI. Fig. 4 shows an example of the clinical importance of considering POI in TB measure-
Fig. 4. Influence of V˙ b-o on the TB parameter tPTEF/tE in healthy infants and infants with CLD (**P B/0.01).
The main goal of this study was to investigate the relation of a POI with different TB measurements. We found that a POI is related mainly to those TB parameters describing the flow profile or the shape of the TBFVL (e.g., the extensively investigated parameter tPTEF/tE). In contrast, conventional TB parameters, which describe only rate and depth of breathing (RR, VT and V˙/E), are relatively independent on POI. The mechanisms explaining the association between POI and TB pattern in newborns are not fully understood. It is not easy to distinguish cause from effect. Most scientists would argue that it is not the POI that causes changes in breathing patterns described by TB parameters but vice versa. Therefore, a strong relationship between the two becomes almost inevitable. Indeed, the braking of expiratory flow together with a shortening of the expiratory time is well recognized as the means by which neonates defend lung volume (Stark et al., 1987). For the clinical use of TB measurements, it is important to know, however, that the POI is associated with significant changes in some TB parameters. In a neonatal lung function laboratory, infants with CLD constitute one of the most important patient groups. Up to now, CLD is the most common respiratory complication in preterm infants requiring prolonged mechanical ventilation (Jobe and Ikegami, 1998). In contrast to expectations, the improved mechanical ventilation and the introduction of surfactant for the treatment of an immature surfactant-deficient lung in preterm infants did not decrease the incidence of CLD, both factors may even have augmented CLD incidence due to the markedly increased survival of the smallest infants (Bancalari and del Moral, 2001).
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
In this study, the differences in TB parameters between both patient groups are in good agreement with published results. Tepper et al. (1986) reported a significantly higher RR in CLD infants compared with controls, but no significant changes in VT. In this study, the unchanged tidal volume related to body weight and the much higher RR in CLD infants compared to healthy neonates explain the significantly higher parameters describing expiratory flow pattern in infants with CLD. In contrast to TB, it is well recognized that at forced expiration, CLD infants show a significant flow limitation due to a poor growth of the airways and the resulting higher peripheral airway resistance (Tepper et al., 1986). In CLD infants of the present study, end-expiratory flow (TEF25) was not reduced. This was probably due to their high RR and the resulting higher expiratory flow rates. The parameters tPTEF, tPTEF/tE as well as VPTEF/VT, which is strongly correlated with tPTEF/tE describe the site of PTEF in the flow signal as well as in the TBFVL. These parameters are frequently used to detect airway obstructions (Martinez et al., 1988; Greenough et al., 1998; Lodrup et al., 1999). However, the association of these parameters with small airway caliber remains speculative and it could not be demonstrated in previously published studies (Aston et al., 1994; Williams et al., 2000). van der Ent et al. (1998) have shown that tPTEF/tE is much more influenced by post-inspiratory muscle activity than by lung mechanics. In this study, we found a slight but statistically significant reduction of tPTEF/tE in CLD infants. The higher PTEF and the shorter tPTEF in CLD infants explain the large differences between both patient groups in the mean initial gas acceleration given by PTEF/tPTEF. An unexpected finding of our study was that conventionally used TB parameters (e.g., RR, VT, V˙/E, mean/V˙/I), which are clearly defined and easy to derive, are relatively independent on POI. Therefore, in contrast to investigations in the past, where tPTEF/tE was the centre of interest (Carlsen et al., 1997; Greenough et al., 1998; Lodrup et al., 1999), TB measurements in neonates should be focused on the evaluation of these conventional
45
parameters measured under standardized conditions (Bates et al., 2000). A distinct POI with a V˙ b-o/ /1/3 PTEF was found in 41% of all infants. This means that a POI is a frequent breathing strategy in healthy and sick infants. Because of this high incidence and its significant influence on some TB parameters, it has to be taken into consideration when analyzing and interpreting TB parameters in newborns. The determination of a POI from TBFVL is subjected to methodological limitations. We have not tested by electromyography whether an increased V˙ b-o indicates a POI in all cases. It is well known from commonly used occlusion tests to measure respiratory mechanics (LeSouef et al., 1984; Gappa et al., 2001) that a V˙ b-o is a good indicator for an incomplete exhalation. In infants with flow braking, an increased end-expiratory flow must not be caused by POI. As shown in Fig. 2, such a pattern was seen in healthy and CLD infants in only 15 and 9%, respectively. In conclusion, a POI affects the interpretation of some TB parameters measured in newborns. The study shows that at this age, TB parameters describing the flow profile may reflect differences in the breathing strategy rather than impaired respiratory functions.
Acknowledgements The authors thank Dr. Mario Schmidt for his support in the development of the software and Mrs. Silke Schmidt for her assistance in respiratory function testing. They also gratefully acknowledge Prof. Colin Morley (Melbourne) for the critical review of the manuscript. This work was supported by the German Ministry for Education and Research, project ‘‘Perinatal Lung’’ (grant 01-ZZ-9511).
References Aston, H., Clarke, J., Silverman, M., 1994. Are tidal breathing indices useful in infant bronchial challenge tests? Pediatr. Pulmonol. 17, 225 /230.
46
G. Schmalisch et al. / Respiratory Physiology & Neurobiology 135 (2003) 39 /46
Bancalari, E., 2001. Changes in the pathogenesis and prevention of chronic lung disease of prematurity. Am. J. Perinatol. 18, 1 /9. Bancalari, E., del Moral, T., 2001. Bronchopulmonary dysplasia and surfactant. Biol. Neonate. 80 (Suppl. S1), 7 /13. Bates, J.H., Schmalisch, G., Filbrun, D., Stocks, J., 2000. Tidal breath analysis for infant pulmonary function testing: ERS/ ATS task force on standards for infant respiratory function testing. Eur. Respir. J. 16, 1180 /1192. Carlsen, K.C.L., Stenzler, A., Carlsen, K.H., 1997. Determinants of tidal flow volume loop indices in neonates and children with and without asthma. Pediatr. Pulmonol. 24, 391 /396. Foitzik, B., Schmidt, M., Windstetter, D., Wauer, R.R., Schmalisch, G., 1998. Leak measurements in spontaneously breathing premature newborns by using the flow-through technique. J. Appl. Physiol. 85, 1187 /1193. Fox, R.E., Kosch, P.C., Feldman, H.A., Stark, A.R., 1988. Control of inspiratory duration in premature infants. J. Appl. Physiol. 64, 2597 /2604. Gappa, M., Colin, A.A., Goetz, I., Stocks, J., 2001. Passive respiratory mechanics: the occlusion techniques. Eur. Respir. J. 17, 141 /148. Greenough, A., Zhang, Y.X., Yuksel, B., Dimitriou, G., 1998. Assessment of prematurely born children at follow-up using a tidal breathing parameter. Physiol. Meas. 19, 111 /116. Hummler, H.D., Gerhardt, T., Gonzalez, A., Bolivar, J., Claure, N., Everett, R., Bancalari, E., 1996. Patienttriggered ventilation in neonates: comparison of a flowand an impedance-triggered system. Am. J. Respir. Crit. Care Med. 154, 1049 /1054. Jobe, A.H., Ikegami, M., 1998. Mechanisms initiating lung injury in the preterm. Early Hum. Dev. 53 (1), 81 /94. Jobe, A.H., Ikegami, M., 2000. Lung development and function in preterm infants in the surfactant treatment era. Annu. Rev. Physiol. 62, 825 /846. Kosch, P.C., Stark, A.R., 1984. Dynamic maintenance of endexpiratory lung volume in full-term infants. J. Appl. Physiol. 57, 1126 /1133. LeSouef, P.N., England, S.J., Bryan, A.C., 1984. Passive respiratory mechanics in newborns and children. Am. Rev. Respir. Dis. 129, 552 /556. Lodrup, C.K., Carlsen, K.H., Nafstad, P., Bakketeig, L., 1999. Perinatal risk factors for recurrent wheeze in early life. Pediatr. Allergy Immunol. 10, 89 /95.
Martinez, F.D., Morgan, W.J., Wright, A.L., Holberg, C.J., Taussig, L.M., 1988. Diminished lung function as a predisposing factor for wheezing respiratory illness in infants. N. Engl. J. Med. 319, 1112 /1117. Nikischin, W., Gerhardt, T., Everett, R., Gonzalez, A., Hummler, H., Bancalari, E., 1996. Patient-triggered ventilation: a comparison of tidal volume and chest wall and abdominal motion as trigger signals. Pediatr. Pulmonol. 22, 28 /34. Patzak, A., Foitzik, B., Mrowka, R., Schmalisch, G., 2001. Time of measurement influences the variability of tidal breathing parameters in healthy and sick infants. Respir. Physiol. 128, 187 /194. Prechtl, H.F., 1974. The behavioural states of the newborn infant (a review). Brain Res. 76, 185 /212. Schmalisch, G., Grubba, K., Wauer, R.R., 1996. Computeraided interpretation of pulmonary parameters in newborn infants. Med. Biol. Eng. Comput. 34, 160 /164. Schmalisch, G., Foitzik, B., Wauer, R.R., Stocks, J., 2001a. Effect of apparatus deadspace on breathing parameters in newborns: ‘‘flow-through’’ versus conventional techniques. Eur. Respir. J. 17, 108 /114. Schmalisch, G., Schmidt, M., Foitzik, B., 2001b. Novel technique to average breathing loops for infant respiratory function testing. Med. Biol. Eng. Comput. 39, 688 /693. Stark, A.R., Cohlan, B.A., Waggener, T.B., Frantz, I.D., Kosch, P.C., 1987. Regulation of end-expiratory lung volume during sleep in premature infants. J. Appl. Physiol. 62, 1117 /1123. Stocks, J., Dezateux, C.A., Jackson, E.A., Hoo, A.F., Costeloe, K.L., Wade, A.M., 1994. Analysis of tidal breathing parameters in infancy: how variable is tPTEF/tE. Am. J. Respir. Crit. Care Med. 150, 1347 /1354. Tepper, R.S., Morgan, W.J., Cota, K., Taussig, L.M., 1986. Expiratory flow limitation in infants with bronchopulmonary dysplasia. J. Pediatr. 109, 1040 /1046. Ueda, Y., Stick, S.M., Hall, G., Sly, P.D., 1999. Control of breathing in infants born to smoking mothers. J. Pediatr. 135, 226 /232. van der Ent, C.K., van der Grinten, C.P.M., Meessen, N.E.L., Luijendijk, S.C.M., Mulder, P.G.H., Bogaard, J.M., 1998. Time to peak tidal expiratory flow and the neuromuscular control of expiration. Eur. Respir. J. 12, 646 /652. Williams, E.M., Madgwick, R.G., Thomson, A.H., Morris, M.J., 2000. Expiratory airflow patterns in children and adults with cystic fibrosis. Chest 117, 1078 /1084.