INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Lung transplantation is an option for selected children with life-threatening pulmonary diseases (1). One hundred forty-six infants and children have received lung transplants at St. Louis Children's Hospital from June 1990 through March 1998, including 37 patients younger than 3 yr of age. These younger patients had a variety of underlying diseases, including bronchopulmonary dysplasia, pulmonary alveolar proteinosis with or without surfactant protein deficiency (4, 5), and pulmonary hypertension that was primary or secondary to congenital heart disease or obstruction of pulmonary veins. The results in recipients younger than 3 yr of age were of particular interest because the lungs of normal children of that age are expected to grow by the addition of several million alveoli (6). This is the first report to document in detail the physiologic development of immature lungs transplanted into immature human recipients. For these investigations we used techniques that have been shown to be useful measures of FRC and airway function in infants.

Our goal was to demonstrate that normal lung growth occurs, at least in some recipients, in the absence of substantial airflow obstruction and despite denervation and immunosuppression regimes that include systemic corticosteroids. In this report, we thus focus on survivors with little or no airflow obstruction.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Infant Pulmonary Function Testing

The infants and children were studied in the pulmonary function laboratory at regular intervals after lung transplantation (3, 6, and 12 mo and twice yearly thereafter). Before pulmonary function testing (PFT), the children were admitted to the ambulatory procedure center for intravenous placement and sedation. A sedation nurse accompanied the child to the pulmonary function laboratory, administered the sedation, monitored the child during testing, and returned the child for recovery to the procedure center. Although chloral hydrate is the sedative-hypnotic of choice for most testing in our infant laboratory, we have had little success with chloral hydrate in transplant recipients. With few exceptions, the infants received secobarbital (5 to 7 mg/kg) and midazolam (0.05 mg/kg). All measurements were done with the infants supine and quietly asleep.

FRC measurements were made via an open-circuit nitrogen washout technique (Sensormedics Corporation, Yorba Linda, CA) (9). Before measurements, the nitrogen analyzer was "peaked" and a two-point calibration was performed. Accuracy and precision of calibration were established before FRC measurements by testing the system several times with a volume from a calibration syringe equal to the child's expected FRC (~ 20 ml/kg). The child was positioned with a small roll under his or her shoulders to slightly extend the neck. Pliable face masks with dead spaces ranging from 32 to 48 ml were placed over the nose and mouth and inspected for possible leaks; leaks could also be identified from the washout curve. Chest excursion was observed, and the child was switched to a circuit with 100% oxygen. Change in percent exhaled nitrogen level decreased to 0%. Between repeated measurements, the child breathed room air for two times his or her washout duration to reestablish alveolar nitrogen levels. We recorded a minimum of three FRC measurements with a goal of obtaining a coefficient of variation of 10% or less; the mean of the three measurements is reported.

After FRC measurements, rapid thoracoabdominal compressions (RTC) were performed to obtain partial expiratory flow-volume curves (10), and flow was recorded as maxFRC/FRC (11). Flow measurements were obtained via a pneumotachometer, and the flow signal was integrated to yield volume. Compressions were accomplished with the Hugger Cart 2605 (Equilibrated BioSystems, Inc., Melville, NY). Studies of forced flow were performed from within the tidal range. Compression pressure within the RTC bag was progressively increased, starting at reservoir pressures of 20 to 25 cm H₂O and increased in increments of 10 cm H₂O. Most often the pressure was taken to the reservoir limit of 100 cm H₂O and then decreased to the pressure where maximal flow at FRC occurred. Criteria for acceptance of curves followed published standards (12): (1) rapid expulsion of expiratory flow with peak flow occurring prior to midpoint of expiratory time, (2) smooth expiratory curve without significant flow transients, especially near FRC, and (3) expiration beyond FRC. The four best curves with highest maxFRC were chosen, and the mean was calculated.

Statistical Analysis

Descriptive data are presented as mean ± standard deviation. Comparisons of continuous variables among groups were done using ANOVA. The relationship between abnormal PFT and evidence for graft rejection or bacterial lower respiratory infection was analyzed by Fisher's exact test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects and Diagnoses

Thirty-seven infants younger than 3 yr of age have received lung transplants at St. Louis Children's Hospital. Serial lung function data have been collected in 23 of the 37 (Table 1); 14 died early in their postoperative course or were otherwise too unstable for serial tests to be done. The underlying diagnoses and the age at transplant for all infants studied are shown in Table 1. All subjects underwent bilateral sequential lung transplantation with cardiopulmonary bypass, and all received cyclosporine A, azathioprine, and prednisone to control graft rejection (2). The dose of prednisone is 0.5 mg/kg/d initially after transplant, with the goal of reducing it to 0.2 mg/kg/d within the first year. Thereafter the dose is 0.10 to 0.15 mg/kg/d.

As a group, the infants without obvious airflow obstruction (Table 2) were growing in body length at a rate slightly slower than normal (0.73 ± 0.18 cm/mo after transplant) (13).

The age at transplant of the infants studied by PFT was 13.2 ± 8.4 mo. The age at transplant of all infants younger than 3 yr of age, including the 14 too unstable to have been studied, was 11.4 ± 8.2 mo.

FRC and Forced Expiratory Flow at FRC

FRC and forced expiratory flow were measured 86 times in 23 subjects (per infant, 3.7 ± 1.7 studies; range, 1 to 6). As noted above, the schedule for PFT specified post-transplant Months 3, 6, 12, 18, 24, and 30 to 36. Thus far, eight of 23 subjects have survived longer than 30 to 36 mo post-transplant. One subject was too unstable at Month 3 for testing; others had long distances to travel and were not studied at the specified interval if they did not return to St. Louis. Nevertheless, the 86 studies reported represent an adherence-to-protocol rate of 89%.

Flow at FRC was measured to detect airflow obstruction; in particular, we wanted to separate infants with near normal airway function from those whose increase in lung volume might have been caused by intrathoracic airway obstruction with air trapping such as occurs in bronchiolitis obliterans or narrowing at the anastomoses.

Of the 23 recipients studied with PFTs, all forced expiratory flow studies were near normal for seven subjects, all studies showed obstruction for seven subjects, and for nine the results were mixed over time, with some studies near normal and others suggesting airflow obstruction. There were 50 studies in 16 subjects where the specific flow at FRC was near normal (> 0.9 FRC/s) (Table 2) (11). The flows at FRC in this group were 1.60 ± 0.60 FRC/s (range, 0.90 to 2.74). There were 36 studies in 16 subjects where the specific flow at FRC was low enough to suggest obstruction (< 0.9 FRC/s). The flows at FRC in this group were 0.34 ± 0.26 FRC/s (range, 0.03 to 0.87).

Bronchoalveolar lavage and transbronchial biopsy was done within 2 d of most PFTs. The diagnosis of graft rejection was made by transbronchial biopsy on 13 occasions in eight infants. Four infants with airflow near normal and nine with probable obstruction had histologic evidence for rejection. Quantitative culture of bronchoalveolar lavage fluid indicated bacterial lower airway infection (> 10⁵ CFU/ml) on five occasions in five infants. The association of apparent obstruction with rejection approached significance (Fisher's exact test, p = 0.06), but there were too few positive BAL cultures to establish an association with obstruction (Fisher's exact test, p > 99).

The intrasubject, intrasession coefficients of variation for all acceptable FRC measurements ranged from 0.30 to 16.9% (3.9 ± 2.8%); only six of 86 studies had coefficients > 8.00% (9). (Those with near normal flow had coefficients of variation of 3.26 ± 2.17%; those presumably obstructed had coefficients of variation of 4.67 ± 3.27%). Because it was likely that small airways obstruction would cause pathologic increases in FRC, FRC data from those with near normal flows and those with presumed obstruction were considered separately. FRC versus body length in the seven infants who always had near normal airflow (FRC/s > 0.9) is shown in Figure 1. The slope of the line is 6.83; the slopes for normal infants in published reports (11, 14) are 5.38 to 7.74. FRC plotted versus length for the nine infants with mixed results is shown in Figure 2; FRC values shown are those recorded when airflow was near normal; the slope of the line is 6.88, nearly the same as in Figure 1. FRC versus body length from infants whose forced expiratory flow studies always suggested obstruction is shown in Figure 3. The slope is 10.89. 

View larger version (14K): [in this window] [in a new window]   Figure 1.   Results form 26 measurements of FRC in seven subjects (Subjects 3, 7, 8, 17, 18, 20, 21). This figure includes results from subjects whose forced expiratory flow studies always showed near normal flow. Slopes for healthy infants (in References 11 and 14) range from 5.38 to 7.74.

View larger version (13K): [in this window] [in a new window]   Figure 2.   Results from 23 measurements of FRC in nine subjects (Subjects 4 to 6, 9 to 12, 15, and 22) at a time when results of forced expiratory flow studies did not suggest significant obstruction, i.e., forced expiratory flow > 0.9 FRC/s. Overall, these subjects had mixed results: 23 of 40 studies were near normal, and the remainder yielded forced expiratory flow < 0.9 FRC/s. Published Slopes from normal infants in References 11 and 14 range from 5.38 to 7.74.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Results from 37 measurements of FRC in 16 subjects at a time when forced expiratory flow studies suggested significant obstruction (flow < 0.9 FRC/s). All studies for subjects 1, 2, 13, 14, 16, 19, 23 suggested obstruction. This graph also includes data from studies showing obstruction from those subjects having mixed results.

Detailed results for subjects with near normal forced expiratory flows at FRC are shown in Table 2. Twelve of 16 subjects in Table 2 had both multiple FRC measurements and near normal flow.

Respresentative FRC data from Table 2 are shown in Figure 4. All FRC data are plotted from four of these 16 subjects; FRC, corrected for body length in centimeters, is plotted against time in months since transplantation (Figure 4). Published normal values (14, 15) are 1.94 to 2.45 ml/cm body length for infants 6 to 25 mo of age. The range of FRC/cm in our studies with near normal airflow was 0.6 to 4.0 ml/cm.

View larger version (23K): [in this window] [in a new window]   Figure 4.   Results from four representative infants from Table 2. (Subjects 3, 8, 9, 10) When FRC was measured in these four infants, they appeared not to have significant airflow obstruction, based on forced expiratory flows > 0.9 FRC/s. Y-axis is FRC in ml, measured by N2 washout, and corrected for body length. X-axis is time in months after transplant. These results demonstrate near normal acquisition of lung volume in these transplant recipients (14, 15) in the absence of significant airflow obstruction.

For the 16 infants with flows at FRC near normal, there was little change in the specific flows at FRC over the study months; the ratios of maximal flow at FRC to FRC varied from only 1.79 ± 0.83 FRC/s (3 mo) to 1.53 ± 0.65 FRC/s (18 mo). The values at each interval were not different (ANOVA, p = 0.91). This suggests that the rates of growth of alveoli and conducting airways were similar (11, 16).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Whether lungs with potential for growth do in fact grow after transplantation into human infants is assumed but has not been established. We used measurements of FRC, a standard method to describe lung growth (9, 11, 14, 15), to study 23 infants. We focused on those infants with near normal expiratory flow because they are less likely to have increases in FRC caused by pathologic processes such as bronchiolitis obliterans. In these recipients we have shown that increases in FRC accompanying somatic growth, when normalized to body length, are comparable to those published for normal subjects.

Massaro and Massaro (17, 18) have recently shown that nursing rat pups given corticosteroids have lungs with emphysematous air spaces and marked reduction in alveolar number. These findings have caused much concern among clinicians using corticosteroids to prevent or treat chronic lung disease in small infants. However, we have demonstrated apparently normal growth in lung volume in a subset of survivors of lung transplantation in infancy in the face of denervation, varying degrees of graft rejection, and immunosuppression with corticosteroids. Massaro and Massaro (17) also believe that, after corticosteroid treatment, reduction in the number of airway attachments to the elastic elements within emphysematous rat lungs causes airflow obstruction because of interdependence of the airways on the weakened elastic elements.

We believe that the increases in FRC are due to increases in number of alveoli in infants with near-normal forced expiratory flow after transplant. This conclusion requires the assumption that, for the lung as a whole, airways with near-normal function serve alveoli whose total surface area is increasing by hyperplasia (alveolarization) rather than by hypertrophy (overdistention). In the absence of other models of disease exhibiting normal airway function but emphysematous air spaces, this assumption seems reasonable. The conclusion that the increases in FRC are due to increases in alveolar number is also buttressed by work done in animals. Quantitative studies of histology by Hislop and colleagues (19) in immature rats that received unilateral transplants showed normal growth of the lungs after 6 mo, with increases in alveolar number and airway diameter both comparable to nontransplant control subjects. A definitive answer about whether growth in lung volume after transplant in humans is due more to hyperplasia or hypertrophy must await comparison of recipients' lungs to size and age-matched control subjects using quantitative morphometry.

During infancy in normal children, the rate of growth of conducting airways varies with respect to lung volume, so that the ratio of maximal flow to lung volume is not constant. Such dysynaptic growth is most marked between birth and 3 mo of age (11, 16). Girls also appear to have a greater rate of physiologic growth of airways than do boys when flows are corrected for lung volume (11). Of our 14 subjects with near normal maxFRC/FRC studied more than once, seven were boys and seven were girls; of the donors, five were girls, with three pairs of lungs going to boys and two to girls; nine donors were boys, with four pairs of lungs going to boys and five to girls. Although our finding in this group was that increases in airway function and alveolar growth are symmetric (ANOVA, p = 0.91), it is based on a relatively small number of studies, and is, thus, preliminary.

Because human lungs can, and apparently often do, show sustained growth when transplanted into infants, our results are a tentative source of encouragement about a group of patients whose clinical courses can be troubling. Indeed, 17 of 37 died, 11 within the first 4 mo after transplant. Nevertheless, 12 of 23 infants studied, without substantial airflow obstruction, had increases in FRC per centimeter body length near that predicted for normal children, suggesting normal lung growth.