METHODS. Subjects. Lung Function

Transcription

1 Worsening of V maxfrc in Infants with Chronic Lung Disease in the First Year of Life A More Favorable Outcome after High-Frequency Oscillation Ventilation Ward Hofhuis, Marianne W. A. Huysman, Els C. van der Wiel, Wim P. J. Holland, Wim C. J. Hop, Govert Brinkhorst, Johan C. de Jongste, and Peter J. F. M. Merkus Department of Pediatrics, Divisions of Respiratory Medicine and Neonatology; Department for Experimental Medical Instrumentation; and Department of Biostatistics, Erasmus University Medical Center/Sophia Children s Hospital, Rotterdam; and Department of Pediatrics, Medical Center Alkmaar, Alkmaar, The Netherlands Little is known about the development of maximal flow at functional residual capacity, a measure of airway patency, in infants with chronic lung disease (CLD). In a follow-up study, we evaluated V maxfrc in very low birth weight infants with CLD, treated with highfrequency oscillation ventilation (HFOV) or conventional mechanical ventilation. In 36 infants with CLD, V maxfrc was evaluated at 6 and/or 12 months corrected age, and the relationship between perinatal factors and lung function was studied. Mean (SD) birth weight and gestational age were 837 (152) g and 26.8 (1.7) weeks, respectively. At 6 and 12 months, mean V maxfrc was significantly below normal. Between 6 and 12 months, there was a mean (95% confidence interval) reduction in V maxfrc (Z score) of 0.5 ( ) (p 0.001). At 12 months, the mean V maxfrc (Z score) was higher for children initially treated with HFOV (n 15), as compared with children treated with conventional mechanical ventilation (n 16): mean (95% confidence interval) difference was 0.6 ( ) (p 0.008). We conclude that very low birth weight infants with CLD have decreased V maxfrc that worsen during the first year of life. Initial treatment with HFOV was associated with a more favorable outcome of V maxfrc at 12 months corrected age. Keywords: neonatal chronic lung disease; prematurity; pulmonary func- tion test; high-frequency oscillation ventilation Chronic lung disease (CLD) is a common sequel of mechanical ventilation and oxygen therapy in prematurely born in- fants (1). Despite advances in prenatal and neonatal care, including antenatal and postnatal steroids, surfactant treatment, and high-frequency oscillation ventilation (HFOV), CLD is still one of the major complications in mechanically ventilated premature infants (2). The overall incidence of CLD has remained high as a result of the increased survival of extremely premature infants, who are most likely to develop CLD (2). Long-term studies show that survivors of CLD have abnormal pulmonary function tests at school age (3, 4), whereas infants who received initial HFOV showed normal lung function at school age (5). Only a few studies evaluated lung function during the first years of life in children with CLD. In young children with CLD, lung function parameters, such as functional residual capacity (FRC p ), compliance, resistance, and conductance, show a gradual improvement to- ward the normal range during the first 3 years of life (6 8). (Received in original form February 19, 2002; accepted in final form August 8, 2002) Supported by an unrestricted grant from GlaxoSmithKline, The Netherlands. Correspondence and requests for reprints should be addressed to Peter J. F. M. Merkus, M.D., Ph.D., Erasmus University Medical Center, Sophia Children s Hospital, Department of Pediatrics, Division of Respiratory Medicine, Room Sp-2469, P.O. Box 2060, 3000 CB, Rotterdam, The Netherlands. merkus@alkg.azr.nl Am J Respir Crit Care Med Vol 166. pp , 2002 DOI: /rccm Internet address: Nevertheless, maximal flow at FRC (V maxfrc ), used as a mea- sure of airway patency, is known to be decreased during the first 2 years of life (6, 8, 9). Due to advances in prenatal and neonatal care, results obtained in the past may not be valid for infants who develop CLD nowadays. There are no recent studies that evaluated V maxfrc during the first year of life in very low birth weight (VLBW) infants with CLD, in the era of surfactant therapy and HFOV. Therefore, we aimed at evaluating V maxfrc at 6 and 12 months corrected age, in a group of VLBW infants with CLD. Furthermore, we studied the relationship between lung function and perinatal patient characteristics. METHODS Subjects A follow-up study was conducted in neonates who developed CLD, born between January 1998 and September All infants were born in or transferred immediately after birth to the Neonatal Intensive Care Unit of the Sophia Children s Hospital. The inclusion criteria were (1) VLBW: birth weight of 1,250 g or less, (2) need for mechanical ventilation from Day 1 for at least 7 days, (3) need for continuous supplemental oxygen at 28 days and/or at 36 weeks gestational age, and (4) chest radiogram at 1 month of age typical for CLD. The exclusion criteria were major congenital anomalies, meconium aspiration, or suspected hypoplasia of the lungs. Artificial ventilation in the Neonatal Intensive Care Unit was administered by conventional mechanical ventilation (CMV) or HFOV. Initial ventilation strategy was not randomized in our study. Preferably, initial HFOV was started in the youngest and smallest infants. This was not always feasible due to the limited avail- ability of HFOV equipment, and hence, initial ventilation strategy was partly determined by chance. When infants developed hyaline mem- brane disease, surfactant (Survanta, 100 mg/kg/dose) was administered. Neonates with severe hyaline membrane disease received additional doses. When infants developed a persistent need for artificial ventilation, treatment also included fluid restriction and diuretics. To wean them off the ventilator, most infants were treated with dexamethasone, administered in a 3-week course starting with a dose of 0.5 mg/kg/ day that was gradually tapered. All infants were age-corrected to a gestational age of 40 weeks. The study was approved by the medical ethical committee of the Erasmus University Medical Center. All parents gave informed consent. Lung Function Lung function measurements were performed at 6 and 12 months corrected age, when the infants were free from acute respiratory symptoms. To prevent the infants from waking up during the measurements, they were sedated with choral hydrate (50 75 mg/kg). FRC p was measured by means of a modified whole body plethysmograph (Jaeger, Würzburg, Germany). Equipment and procedures were in accordance with recently published guidelines, in which the FRC p measurement is described in detail (10). The mean FRC p of three to five technically acceptable measurements was expressed as Z score (10). V maxfrc was assessed using the end-tidal rapid thoracoabdominal compression technique (RTC) (Custom-made equipment; Department for Experimental Medical In-

2 1540 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL TABLE 1. ANTHROPOMETRIC DATA Infants Measured Infants Measured Total Group at 6 mo at 12 mo (n 36) (n 28) (n 31) GA, wk 26.8 (1.7) 26.9 (1.7) 26.9 (1.7) Birth weight, g 837 (152) 826 (156) 852 (156) Birth weight, Z score* 1.2 (1.3) 1.4 (1.1) 1.2 (1.4) Males Maternal steroids Tocolyses PROM PDA Surfactant-treated newborns Dexamethasone-treated newborns Duration of dexamethasone treatment, d 23 (15) 21 (10) 25 (16) Initial HFOV Duration of ventilation, d 27 (13) 25 (10) 27 (14) Duration of oxygen dependence, d 151 (161) 150 (174) 166 (169) Oxygen dependence at 28 d Oxygen dependence at GA 36 wk Definition of abbreviations: GA gestational age; HFOV high-frequency oscillation ventilation; PDA persistant ductus arteriosus; PROM premature rupture of membranes. Data given are number of infants or mean (SD). Shown are the total group and the subgroups of infants for whom measurements were made at 6 and 12 months corrected age. Twenty-three infants completed both measurements. * Reference values by Usher and McLean (22). strumentation, Erasmus University Medical Center, Rotterdam, The ments were failure to sleep during the procedure (n 6), airway Netherlands). Equipment and procedures were in accordance with re- infections (n 5), and loss to follow-up (n 2). Anthropometric cently published guidelines, in which the rapid thoracoabdominal com- data of the total cohort of 36 infants and of the subgroups of pression technique is described in detail (11). The mean V maxfrc of three 28 infants measured at 6 months and 31 infants measured at 12 to five technically acceptable measurements was expressed as Z score months are shown in Table 1. The first and second lung function according to Sly and coworkers (11) and Tepper and Reister (12). measurements were performed at mean (SD) corrected ages of Analysis 6.2 (0.9) months and 12.6 (1.1) months, respectively. The results Lung function at the first and second measurements was compared using of the FRC p and the V maxfrc measurements are shown in Table 2. mixed-model analysis of variance (SAS, PROC MIXED). Between Mean (SEM) FRC p values in Z score at the first and second the groups initially treated with HFOV or CMV, lung function and measurements were 1.2 (0.3) and 0.6 (0.2), respectively. anthropometric data were compared using independent-samples t tests. Mean (SEM) V maxfrc in Z score was significantly below zero Comparison of percentages was done using Fisher s exact test. Where (normal value) at the first and second measurements: 1.7 (0.1) applicable, the difference in lung function was evaluated using paired and 2.2 (0.1), respectively (Table 2, Figure 1). Between the Student s t test. The influence of various perinatal variables on the level two measurements, there was a mean (95% confidence interval) of lung function was evaluated by multiple regression analyses. The change of V maxfrc in Z score of 0.5 ( 0.7 to 0.2) (p 0.001). significance level was set at a p value of less than When V maxfrc in Z score was calculated using normative data RESULTS by Tepper and Reister (12), similar results were seen: the mean (SEM) V maxfrc (Z score) values at the first and second measurements A cohort of 36 white infants was enrolled. Lung function was were 1.6 (0.1) and 2.0 (0.1), respectively (mean [95% measured in 28 infants at 6 months and in 31 infants at 12 months. confidence interval] change of V maxfrc in Z score of 0.4 [ 0.7 In 23 infants, lung function was measured both at 6 and 12 to 0.1], p 0.006). months corrected age. Reasons for not completing both measure- At 12 months, the mean (SEM) V maxfrc in Z score was better TABLE 2. LUNG FUNCTION DURING FIRST YEAR OF LIFE IN INFANTS WITH CHRONIC LUNG DISEASE Measurement 1 Measurement 2 (n 28) (n 31) Mean Difference Mean SEM Mean SEM (95% CI) FRC p, ml/kg ( 1.3 to 5.0) FRC p, Z score* ( 0.2 to 1.4) V maxfrc, ml/s (26.9 to 72.4) V maxfrc, Z score ( 0.7 to 0.2) Definition of abbreviations: CI confidence interval; FRC p functional residual capacity; V maxfrc maximal flow at FRC. At Measurements 1 and 2, the mean SD corrected ages were and months, respectively. * Reference equation by Stocks and coworkers (10). Reference equation by Sly and coworkers (11). p

3 Hofhuis, Huysman, van der Wiel, et al.: Expiratory Flow in Infants with CLD 1541 TABLE 3. LUNG FUNCTION AT 12 MONTHS CORRECTED AGE, AFTER INITIAL HIGH-FREQUENCY OSCILLATION VENTILATION OR CONVENTIONAL MECHANICAL VENTILATION HFOV CMV Mean Difference (n 15) (n 16) (95% CI) FRC p, ml/kg NS FRC p, Z score* NS V maxfrc, ml/s (24.9 to 92.0) V maxfrc, Z score (0.2 to 1.0) Definition of abbreviations: CI confidence interval; CMV conventional mechanical ventilation; FRC p functional residual capacity; HFOV high-frequency oscillation ventilation; V maxfrc maximal flow at FRC. * Reference equation by Stocks and coworkers (10). Reference equation by Sly and coworkers (11). Figure 1. Lung function data of 36 infants with CLD. Mean V p maxfrc is p expressed in Z score according to Sly and coworkers (11). The first (n 28) and second (n 31) measurements were done at mean (SD) corrected ages of 6.2 (0.9) months and 12.6 (1.1) months, respectively. Twenty-three infants completed both measurements (connected data points). between 6 and 12 months precisely, individual mean V maxfrc values were inter- or extrapolated linearly to values at exactly 6 and 12 months corrected age. At 6 and 12 months, the mean (SEM) of these adjusted V maxfrc (Z score) values were 1.7 (0.1) in the group that received initial HFOV (n 15) as compared and 2.1 (0.1), respectively (mean [95% confidence interval] with the group that initially received CMV (n 16): 1.9 (0.2) change: 0.4 [ 0.7 to 0.1], p 0.006). At 12 months, the mean and 2.5 (0.1), respectively (mean [95% confidence interval] (SEM) of the adjusted V maxfrc in Z score was better in the group difference: 0.6 [ ], p 0.008) (Table 3). The distributions that received initial HFOV (n 12) as compared with the group of perinatal factors did not differ between these two groups, that received CMV (n 11): 1.9 (0.2) and 2.4 (0.1), respecexcept for birth weight (in grams) and requirement of surfactant tively (mean [95% confidence interval] difference: 0.6 [ ], therapy (Table 4). After allowing for the potential confounders p 0.014) (Figure 2). (days on ventilation, gestational age, and birth weight) using multiple regression analyses, this difference remained significant DISCUSSION (p 0.038). However, when both ventilation groups were compared with adjustment for number of surfactant dosages, the In a follow-up study, we evaluated lung function in a cohort difference in mean V maxfrc (Z score) at 12 months lost signifi- of 36 VLBW infants with CLD, during the first year of life. cance (p 0.085). Furthermore, we studied the relationship between lung function Similar results were seen within the subgroup of 23 infants and perinatal patient characteristics. During the first year of who completed both measurements. To study the difference life, mean V maxfrc was below reference values and showed a TABLE 4. ANTHROPOMETRIC DATA OF 31 INFANTS MEASURED AT 12 MONTHS CORRECTED AGE, SUBGROUPED BY INITIAL VENTILATION TREATMENT Initial HFOV Initial CMV (n 15) (n 16) GA, wk 26.5 (1.7) 27.2 (1.8) Birth weight, g 778 (135) 921 (146) Birth weight, Z score* 1.5 (1.6) 0.9 (1.1) Males 12 7 Maternal steroids Tocolyses 12 9 PROM 3 2 PDA Surfactant-treated newborns 9 15 Number of surfactant doses 1 (0 2) 2 (0 4) Dexamethasone-treated newborns Duration of dexamethasone treatment, d 24 (11) 25 (20) Duration of ventilation, d 28 (10) 26 (17) Duration of oxygen dependence, d 127 (128) 202 (197) Oxygen dependence at 28 d Oxygen dependence at GA 36 wk Definition of abbreviations: CMV conventional mechanical ventilation; GA gestational age; HFOV high-frequency oscillation ventilation; PDA persistent ductus arteriosus; PROM premature rupture of membranes. Data given are number of infants, mean (SD) or median (range). * Reference values by Usher and McLean (22). p p p

4 1542 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL combination with surfactant therapy ameliorates acute neonatal lung injury that predisposes some preterm infants to develop CLD (5). The HIFI study group (21) found that the use of HFOV, in comparison with CMV, did not improve V maxfrc at 9 months corrected age. In our study, the V maxfrc at 12 months was significantly better in the group initially treated with HFOV, compared with the group initially managed with CMV. This discrepancy could be explained by the difference in timing of measurement or by the fact that in our study, HFOV was used as initial therapy. Our data suggest that, in VLBW infants, initial treatment with HFOV is associated with a more favorable development of V maxfrc at 12 months corrected age. This finding provides further suggestive evidence that initial HFOV combined with surfactant therapy reduces acute neonatal lung injury (5). Initial ventilation treatment was not intentionally randomized in our study, and therefore this association cannot be considered causal. Nevertheless, the HFOV and CMV groups were not different by any perinatal patient characteristic, except for a small difference in birth weight in grams, but not in Z score, and number of surfactant doses. The difference in birth weight does not explain our finding, as the lower birth weight of the HFOV group would unfavorably affect lung function, whereas we found better results after HFOV. Fewer doses of surfactant were given to the infants who were initially ventilated with HFOV, as compared with CMV. This may reflect reduced respi- ratory distress after HFOV. We regard the number of surfactant doses not as a confounder but as a possible first positive outcome of HFOV. In summary, VLBW infants with CLD, born in the era of surfactant therapy and HFOV, show a worsening of decreased V maxfrc during the first year of life. Initial treatment with HFOV was associated with a more favorable development of V maxfrc at 12 months corrected age. This finding supports the hypothesis that initial treatment with HFOV in premature neonates prone to develop CLD leads to less airway damage and better medium- term outcome. Figure 2. Effect of first intention HFOV on V maxfrc. Mean V maxfrc in Z score (11) in the subgroup of 23 infants who completed both lung function measurements at 6 and 12 months. Individual mean V maxfrc values were inter- or extrapolated to values at exactly 6 and 12 months corrected age. Infants treated with first intention HFOV (open symbols, n 12) are compared with infants treated with CMV (closed symbols, n 11). Error bars represent SEM. significant worsening between 6 and 12 months corrected age. At 12 months, mean V maxfrc was significantly better in the initial HFOV-treated group, as compared with the group treated with CMV. To our knowledge, this is the first study on growth of airway function during the first year of life in VLBW infants with CLD, which also addresses a possible relationship with HFOV. Tepper and coworkers (6) and Iles and Edmunds (9) also found de- References creased V maxfrc during the first year of life. However, due to 1. Northway WH. Bronchopulmonary dysplasia: twenty-five years later. the survival of younger and smaller infants and differences in Pediatrics 1992;89: treatment modalities, our study population cannot be compared 2. Eber E, Zach MS. Long term sequelae of bronchopulmonary dysplasia with the population studied by Tepper and coworkers (6). Iles (chronic lung disease of infancy). Thorax 2001;56: and Edmunds (9) studied a population more comparable to our 3. Jacob SV, Coates AL, Lands LC, MacNeish CF, Riley SP, Hornby L, population, but no information about treatment modalities was Outerbridge EW, Davis GM, Williams RL. Long-term pulmonary sequelae of severe bronchopulmonary dysplasia. J Pediatr 1998;133: provided. The decreasing V maxfrc may reflect abnormal func tional or anatomic development of the airways (6), which is 4. Gross SJ, Iannuzzi DM, Kveselis DA, Anbar RD. Effect of preterm birth consistent with pathologic findings (13). This could explain the on pulmonary function at school age: a prospective controlled study. abnormal pulmonary function tests in preterm-born children J Pediatr 1998;133: with CLD at school age (3, 4). Alternatively, worsening of airway 5. Gerstmann DR, Wood K, Miller A, Steffen M, Ogden B, Stoddard RA, patency may be due to airway damage and dysfunction of periphventilation for neonatal respiratory distress syndrome. Pediatrics 2001; Minton SD. Childhood outcome after early high-frequency oscillatory eral airways (14), and central airway damage and collapse during 108: dynamic compression (15). Factors such as thickened airway 6. Tepper RS, Morgan WJ, Cota K, Taussig LM. Expiratory flow limitation walls, increased smooth muscle layer, disturbed development of in infants with bronchopulmonary dysplasia. J Pediatr 1986;109:1040 airway size and/or airway compliance (16), or altered alveolar architecture may also play a role here (17). Furthermore, the 7. Gerhardt T, Hehre D, Feller R, Reifenberg L, Bancalari E. Serial deter- relative decline of V mination of pulmonary function in infants with chronic lung disease. maxfrc during the first year of life was irre- J Pediatr 1987;110: spective of the reference equation used (11, 12). The FRC p was 8. Baraldi E, Filippone M, Trevisanuto D, Zanardo V, Zacchello F. Pulmowithin the normal range at 6 months and demonstrated a trend nary function until two years of life in infants with bronchopulmonary to normalization at 12 months of age. This is consistent with dysplasia. Am J Respir Crit Care Med 1997;155: previous reports (6 8, 18). With no apparent decline of the mean 9. Iles R, Edmunds AT. Assessment of pulmonary function in resolving FRC p between 6 and 12 months, the change in FRC p cannot chronic lung disease of prematurity. Arch Dis Child Fetal Neonatal explain the reduction in V Ed 1997;76:F113 F117. maxfrc (19). 10. Stocks J, Godfrey S, Beardsmore C, Bar-Yishay E, Castile R. Plethysmo- First intention HFOV is associated with a shorter time of graphic measurements of lung volume and airway resistance. ERS/ ventilator dependency and oxygen dependency in VLBW infants ATS Task Force on Standards for Infant Respiratory Function Testing: with respiratory distress syndrome (20). Furthermore, it is specu- European Respiratory Society/American Thoracic Society. Eur Respir lated that early HFOV used with a lung recruitment strategy in J 2001;17:

5 Hofhuis, Huysman, van der Wiel, et al.: Expiratory Flow in Infants with CLD Sly PD, Tepper R, Henschen M, Gappa M, Stocks J. Tidal forced expira- Pulmonary function in infants with neonatal chronic lung disease with tions. Eur Respir J 2000;16: or without hyaline membrane disease at birth. Eur Respir J 1997;10: 12. Tepper RS, Reister T. Forced expiratory flows and lung volumes in normal infants. Pediatr Pulmonol 1993;15: Modl M, Eber E, Weinhandl E, Gruber W, Zach MS. Assessment of 13. Bonikos DS, Bensch KG, Northway WH Jr, Edwards DK. Bronchopulson bronchodilator responsiveness in infants with bronchiolitis: a comparimonary dysplasia: the pulmonary pathologic sequel of necrotizing of the tidal and the raised volume rapid thoracoabdominal com- bronchiolitis and pulmonary fibrosis. Hum Pathol 1976;7: pression technique. Am J Respir Crit Care Med 2000;161: Taussig LM, Landau LI, Godfrey S, Arad I. Determinants of forced 20. Rimensberger PC, Beghetti M, Hanquinet S, Berner M. First intention expiratory flows in newborn infants. J Appl Physiol 1982;53: high-frequency oscillation with early lung volume optimization im- proves pulmonary outcome in very low birth weight infants with respi- 15. Panitch HB, Keklikian EN, Motley RA, Wolfson MR, Schidlow DV. ratory distress syndrome. Pediatrics 2000;105: Effect of altering smooth muscle tone on maximal expiratory flows in 21. The HIFI Study Group. High-frequency oscillatory ventilation compared patients with tracheomalacia. Pediatr Pulmonol 1990;9: with conventional intermittent mechanical ventilation in the treatment 16. Tiddens HA, Hofhuis W, Bogaard JM, Hop WC, de Bruin H, Willems of respiratory failure in preterm infants: neurodevelopmental status LN, de Jongste JC. Compliance, hysteresis, and collapsibility of human at 16 to 24 months of post-term age. J Pediatr 1990;117(6): small airways. Am J Respir Crit Care Med 1999;160: Usher R, McLean F. Intrauterine growth of live-born Caucasian infants 17. Hjalmarson O, Sandberg K. Abnormal lung function in healthy preterm at sea level: standards obtained from measurements in 7 dimensions infants. Am J Respir Crit Care Med 2002;165: of infants born between 25 and 44 weeks of gestation. J Pediatr 1969; 18. Merth IT, de Winter JP, Zonderland HM, Borsboom GJ, Quanjer PH. 74: