© 2002 American Thoracic Society
Surfactant Synthesis and Kinetics in Infants with Congenital Diaphragmatic HerniaDepartment of Pediatrics, Division of Pediatric Surgery, and Institute of Internal Medicine, University of Padova, Padova, Italy; Department of Pediatrics, Erasmus University, Rotterdam, The Netherlands; and Institute of Child Health and Great Ormond Street Hospital, London, United Kingdom Correspondence and requests for reprints should be addressed to Virgilio P. Carnielli, M.D., Ph.D., Senior Lecturer, Pediatric Nutrition, Institute of Child Health, Great Ormond Street Hospital, 30 Guilford Street, MRC CNRC Room W.4.03, London WC1N 1EH, UK. E-mail: v.carnielli{at}ich.ucl.ac.uk
In animal models of congenital diaphragmatic hernia (CDH), surfactant deficiency contributes to the pathophysiology of the disease; however, information on CDH in humans is limited. We compared surfactant disaturated phosphatidylcholine (DSPC) synthesis and metabolism, by stable isotope technology, in newborn infants with CDH and in control subjects. DSPC amount, total proteins, and surfactant proteinA (SP-A) from tracheal aspirates were also measured. DSPC and SP-A were significantly lower in 14 infants with CDH than in the eight control subjects. Mean DSPC was 2.3 ± 1.3 mg/ml of epithelial lining fluid (ELF) in infants with CDH and 4.6 ± 1.5 mg/ml of ELF in control subjects (p = 0.001). Mean SP-A in infants with CDH and in control subjects was 16.2 ± 9.3 and 61.2 ± 30.6 µg/ml of ELF, respectively (p = 0.03). DSPC kinetics was measured in 12 of 14 infants with CDH and in 5 of 8 control subjects. Secretion time was 8.3 ± 5.5 and 8.5 ± 2.5 hours and peak time 51.9 ± 15.2 and 51 ± 13 hours in infants with CDH and in control subjects, respectively. Fractional synthesis rate was not different for infants with CDH and control subjects (p = 0.4). In conclusion, surfactant DSPC synthesis and kinetics were not significantly deranged in infants with CDH compared with control subjects. Other factors, such as lower surface area or increased DSPC catabolism, may contribute to surfactant pool alteration in CDH.
Key Words: surfactant congenital diaphragmatic hernia stable isotopes
Congenital diaphragmatic hernia (CDH) remains an unsolved problem, with high morbidity and mortality despite recent advances in respiratory support. Pulmonary hypoplasia and persistent pulmonary hypertension are the landmarks of the disease, but both morphologic and biochemical lung immaturity are also present. Evidence of biochemical lung immaturity comes mainly from two experimental models: surgical CDH in fetal lambs (1), and chemical CDH in rats induced by 2,4-dichlorophenyl-p-nitrophenyl ether (TOK or nitrofen) during pregnancy (2). In these animal models, lung biochemical immaturity was assessed by morphometric analysis, glycogen concentrations, and disaturated phosphatidylcholine (DSPC) lung content (3) as well as lecithin/sphingomyelin ratio, phosphatidylglycerol concentration (4, 5), and surfactant proteinA (SP-A) concentration in the amniotic fluid (6). In bronchoalveolar lavage fluid of lambs with CDH, the total amount of phospholipids and the percentage of phosphatidylcholine were significantly lower compared with control animals (7), whereas lecithin/sphingomyelin ratio and phosphatidylglycerol concentrations in the amniotic fluids of the same animals were normal. Furthermore, decreased DSPC concentrations have been observed in the lungs of rat pups with CDH (4). A more recent study suggests that differences in surfactant specific protein composition and content are related to lung immaturity, and that they disappeared by the end of gestation in rat pups with CDH (8). In humans with CDH, biochemical immaturity has been assessed in nonsurvivor CDH lungs (9), in amniotic fluid, and, more recently, in bronchoalveolar lavage (10). In lung tissue of infants with CDH, the DSPC concentration was lower in most of the affected lungs compared with age-matched control subjects (9). In the amniotic fluid of infants with CDH, decreased values of lecithin/sphingomyelin ratios, DSPC concentrations, and SP-A concentrations have been reported (11, 12), whereas other researchers found normal lecithin/sphingomyelin ratios and phosphatidyglycerol (7) in the amniotic fluid of the same animal model. In bronchoalveolar lavage of infants with CDH, a recent study by Ijsselstijn and coworkers reported concentrations of phosphatidylcholine, phosphatidylglycerol, and lecithin/sphingomyelin ratios similar to those of age-matched control infants (10), suggesting that a primary surfactant deficiency is unlikely in infants with CDH. Moreover, phosphatidylcholine concentrations in bronchoalveolar lavage seemed not to decrease over time in the same group of infants, although data showed a wide range of variation. In the study by Ijsselstijn and coworkers, however, the control group had severe respiratory failure, which could also have led to secondary surfactant deficiency (10). Recently, we developed a method to measure surfactant synthesis and turnover in vivo in human infants by means of stable isotopes (13). We applied this new technique to study whether the endogenous DSPC surfactant synthesis is impaired in newborn infants with CDH compared with newborn infants with normal lungs.
Patients and Study Design Surfactant kinetics, tracheal aspirate (TA) DSPC, and SP-A were studied in 14 infants with CDH and in 8 newborns with normal lungs. All patients were admitted to the Neonatal Intensive Care Unit, Padova, Italy. Inclusion criteria were the following: (1) respiratory failure due to CDH or upper airway obstruction, or abdominal wall defects and normal lungs; (2) arterial and venous lines for clinical monitoring; (3) written informed consent from parents; (4) no exogenous surfactant administered; and (5) stable cardiorespiratory conditions for at least 18 hours after birth. Blood pressure and pre- and postductal arterial saturations were monitored continuously, and blood gases were checked at least every 8 hours in infants with CDH and once a day in control subjects. The study was approved by the Ethics Committee of the Azienda Ospedaliera di Padova, and written informed consent was obtained from both parents.
Surfactant Kinetics Enrichments were expressed as atom percent excess (14). No amniotic fluid or TA was collected at birth.
Protein Assay
Urea Assay
SP-A
Surfactant Kinetic Parameters Data are presented as mean ± SD. Mean DSPC and SP-A values were calculated by obtaining the average of all time points for each patient and then the means for the patients in each group. Comparisons were done by a nonparametric t test. The level of significance was p < 0.05. Correlations were calculated within the CDH group with Pearson correlation.
We studied 14 infants with CDH and 8 control subjects. Clinical characteristics of the patients with CDH are shown in Table 1. Twelve of 14 infants with CDH and 5 of 8 control subjects received isotope infusions long enough to obtain data on surfactant kinetics. The two groups were matched for weight (p = 0.17), gestational age (p = 0.28), and postnatal age, although the last one was close to statistical significance (p = 0.06). Five infants with CDH died before hospital discharge, two of them 16 and 24 hours after the start of the study. No DSPC kinetic data were obtained from these two infants. Eleven infants with CDH were ventilated on high frequency oscillatory ventilation (HFOV) during the study, and three infants with CDH and all control subjects were on conventional ventilation. None of the infants received extra corporeal membrane oxygenation (ECMO). Three of the eight control subjects did not receive any isotope infusion because parent consent was not obtained. No systemic hypotension or pre- or postductal desaturation differences were recorded during the study period in any patients.
Table 2 reports mean DSPC, proteins, and SP-A recovered from TA as values corrected per milliliter of TA volume and per milliliter of ELF. Significantly lower amounts of TA DSPC and SP-A were found in infants with CDH. Mean TA DSPC values were 2.3 ± 1.3 and 4.6 ± 1.5 mg/ml ELF (p = 0.03) and mean SP-A values were 16.2 ± 9.3 and 61.2 ± 30.6 µg/ml ELF in infants with CDH and in control subjects, respectively (p = 0.001). Total TA protein content was higher in infants with CDH, although not significant (9.6 ± 6.1 mg/ml ELF in infants with CDH versus 6.1 ± 2.3 mg/ml ELF in control subjects, p = 0.07). The changes of DSPC, proteins, and SP-A over time were calculated from the slope of the TA DSPC, total protein, and SP-A amounts over the study period. Table 2 reports mean ± SD regression coefficients in infants with CDH and in the control subjects. Notably, the SP-A amount tended to increase over time in all control subjects, whereas it decreased over time (negative regression coefficient) in all infants with CDH, except one. Due to the large standard deviation, the difference was not significant. Figure 1 reports all TA DSPC and protein values obtained from infants with CDH (Panels A and B) and control infants (Panels C and D). Figure 2 shows SP-A in infants with CDH and in control subjects over the study period.
In all infants, the isotopic enrichments of plasma free fatty acids (FFA)-PA reached steady state in 5.5 to 24 hours during the isotope infusion. Plasma FFA and free plasma PA concentrations were stable during the study period (not shown). A significant incorporation of the intravenously infused labeled-PA was measurable in the surfactant DSPC from all patients. Kinetic data calculated for each individual patient are shown in Figure 3. Surfactant DSPC HL could be reliably calculated for 4 of 5 control subjects and for 10 of 12 infants with CDH. Patients with missing values were extubated shortly after 60 hours from the start of the study.
FSR for PA was 22.0 ± 17.1% per day in infants with CDH and 17.0 ± 11.0% per day in control subjects. ST for PA was similar in patients with CDH and in control subjects, 8.3 ± 5.5 and 8.5 ± 2.5 hours, respectively. PT of surfactant DSPC was reached after 51.9 ± 9.5 and 51.0 ± 9.5 hours from the start of infusion, for infants with CDH and control subjects, respectively. DSPC HL was longer in the CDH group, but the difference was not significant (59.0 ± 21.0 versus 43.0 ± 6.0 hours, p = 0.2). We found a significant positive correlation between the FSR of surfactant DSPC and the changes in DSPC amount from TA over the study period in infants with CDH (R = 0.72, p = 0.008). No correlation was found between DSPC kinetic parameters and PaO2/FIO2 ratio, or alveolararterial oxygen gradient, or mean airway pressure, or oxygenation index in the CDH group. Three infants with CDH who received isotope infusion died after the end of the study after 19.7 ± 9.5 days of life. The mean FSR, ST, PT, and HL were 0.16 ± 0.05%, 6.0 ± 7.3 hours, 50.0 ± 3.5 hours, and 53.7 ± 7.6 hours, respectively, whereas in the CDH survivors, they were 0.24 ± 0.2%, 9.1 ± 5.0 hours, 52.6 ± 18.1 hours, and 82.4 ± 45.4 hours, respectively.
This study measured for the first time surfactant DSPC kinetics in newborn infants with CDH. Data on DSPC, total protein, and SP-A content in TA of newborns with CDH and with normal lungs, and correlation with kinetic data are also reported. Infants with CDH exhibited significantly lower DSPC and SP-A in TA. These values decreased further during the study period compared with age-matched infants with no lung disease. These findings are suggestive of a surfactant deficiency in our patients with CDH, which could reflect a smaller alveolar surface area in the affected lungs or be due to decreased synthesis or increased catabolism of alveolar surfactant (24). Although it is very likely that these differences were due to the lung condition (infants with CDH versus control subjects), in principle it cannot be excluded that differences in ventilator support and inspired oxygen may have contributed to the differences found at study time 0, which on average occurred at 31 hours of life. Our control group had no apparent lung disease and did require minimal amount of supplemental oxygen while on mechanical ventilation, and thus it represented the best possible control group of mechanically ventilated newborn infants (25). As shown by others (26), we also found that the protein content in TA was higher in infants with CDH during the first few days of life compared with that in control subjects, although the difference in our study was not statistically significant (1). Higher protein content in TA could result from lung immaturity, ventilation, and oxygen-induced lung injury. Although the protein content found in TA from infants with CDH was not markedly elevated, it is possible that proteins contributed to a reduction of surfactant activity (1). SP-A amount in TA of infants with CDH was approximately 25% of the control values during the study period. Low amounts of SP-A have been reported in amniotic fluid of human pregnancies (12), and low SP-A mRNA amounts have been shown in lungs of nitrofen-induced CDH rat model at different gestational ages (6), and in term lambs with surgically induced CDH (27). The mechanisms responsible for a decrease in expression of SP-A in CDH Type II cells may be related, at least in part, to the chronic compression of the lungs, which may limit the growth-promoting stimulus of lung stretch and fetal breathing movements (28). After birth, the low alveolar surface can contribute to the reduced DSPC and SP-A pool size. The novel information of the study is that the endogenous surfactant synthesis as expressed as DSPC FSR from PA was not statistically different between infants with CDH and control subjects (22% per day in CDH versus 17% per day in control subjects, p = 0.16). We also found that DSPC FSR positively correlated with TA amount of DSPC over the study period (p = 0.008, R = 0.7). The meaning of the DSPC FSR as compared with the absolute synthesis rate has been discussed at length elsewhere (29). It is notable that a decreased DSPC pool size may result in a higher than expected DSPC FSR. In this study, we have no information on the DSPC pool size in infants with CDH. We assumed a steady state condition, which implies that the amount of DSPC removed equaled the DSPC replacement on the alveolar surface. Given that the mean DSPC FSR in the CDH group was greater than in control subjects by about 30%, a similar ASR could be expected if lung DSPC pool size in the CDH group were to be reduced down to 30% less than in control subjects. On the other hand, if the mean DSPC pool size were reduced by more than 30% in the CDH group, we could have hypothesized a reduced DSPC synthesis and/or secretion based on 13C-PA enrichments in DSPC TA. In animal studies, steady state can be proved by direct measurement of DSPC pool size at different time points during the study period. This is not feasible in vivo in human infants. We therefore provided the following three points as indirect evidence of steady state: (1) we plotted ELF DSPC for each infant over the study period, and all the obtained slopes were not different from zero (data not shown); (2) all subjects were studied beyond the period of the rapid changes of surfactant pools that occur in the first 24 hours of life (23); (3) all study infants were on a constant degree of ventilatory support, indicating a stable clinical and respiratory condition. Despite the previous three considerations, if a steady state were not present in our sick, but clinically stable, infants with CDH, even a pool size change of 10% per day (which would lead to a total DSPC pool loss in about 15 days) would result in an overestimation of DSPC FSR value of about 15%. This would not change the main result of our study. A 22% DSPC FSR in the CDH group represents the mean of widely scattered values. We did not find any correlation with the degree of lung hypoplasia, although the small number of infants did not allow us to draw any conclusion. It is also to be kept in mind that CDH lungs may be under "maximal stimulation" for surfactant synthesis, whereas control lungs were not, and therefore, an impaired DSPC synthesis or secretion cannot be excluded with the present study design. We are in the process of studying whether or not neonates with congenitally normal lungs but with conditions such as meconium aspiration syndrome (MAS) of neonatal pneumonia and requiring significant ventilator support and high fractions of inspired oxygen are able to mount a much greater DSPC synthesis than infants with CDH on comparable ventilator settings. It is notable, however, that DSPC ST, PT, and HL are similar in infants with CDH and in control newborns, in support of a comparable rate of synthesis and secretion in the two groups. Other factors that affect incorporation of PA in surfactant DSPC have to be considered, such as the effect of pulmonary hypertension and ventilator modes. Severe pulmonary hypertension with right-to-left shunt through ductus Botalli may be associated with CDH. Under these circumstances, less blood flow passes through the lungs, and therefore, fewer surfactant metabolic precursors may reach the Type II cells. In all our infants, we could not show any clinical sign of pulmonary hypertension during the isotope infusion. Therefore, we assumed that the label was adequately and equally provided to both study groups. Mechanical ventilation per se does affect DSPC biosynthesis. It has been reported that incorporation of injected radioactively labeled PA into DSPC alveolar fraction increased significantly in a group of ventilated rabbits compared with spontaneously breathing animals (30, 31). In the present study, all infants were on mechanical ventilation, and 79% of infants with CDH were on HFOV. We have recently studied surfactant phosphatidylcholine synthesis in 19 preterm infants, randomized either to HFOV or to conventional ventilation, receiving a continuous infusion of U-13C glucose. We could not demonstrate any difference on phosphatidylcholine FSR, or any difference on the other phosphatidylcholine kinetic parameters in the two randomized groups. Whether this might also be applied to the CDH hypoplastic lungs is still to be determined. It is also important to note that the TA sampling site in our study is just below the endotracheal tube and above the carina, meaning that we studied the contribution of both lungs to surfactant components and kinetics. Therefore, we were not able to ascertain if in infants with CDH the most affected lung had different DSPC kinetics compared with the healthier lung. Few studies are available on surfactant synthesis in CDH animal models, and they show contradictory results. In fetal lamb with surgically induced CDH, the rate of synthesis by isolated Type II cells obtained from CDH lambs was decreased, although the difference did not reach statistical significance (1). In cultures of Type II cells from rat fetuses with severe CDH, Zimmermann and colleagues found a reduced activity of the phosphocholine cytidiltransferase, which regulates the de novo synthesis of phosphatidylcholine (32). Stone and Manson studied the effect of nitrofen on the lung development of normal rat fetuses. They showed that nitrofen did not affect the DSPC synthesis, as the total phospholipid content, choline incorporation, and phosphocholine cytidiltransferase activity were similar to those in control animals (33). Moreover, when each lung in a lung model of CDH was examined, Wilcox and coworkers found that the biosynthetic pathway of Type II cells isolated from CDH lungs was lower, not only compared with control subjects but also compared with the contralateral CDH lungs (26). The lack of biochemical surfactant differences between the two CDH lungs reported in this study has been explained using radiolabeled exogenous surfactant injected in the contralateral lung, as an in utero mixing of lung fluid (26). In conclusion, this study showed moderately but significantly reduced amounts of DSPC in CDH and also a marked reduction of SP-A from TA of CDH infants throughout the study period. Nonetheless, by using a novel method we found comparable rates of endogenous DSPC synthesis in infants with CDH and in control neonates. We speculate that surfactant alteration in CDH may be more related to enhanced catabolism than impaired synthesis or that other factors such as surfactant proteins or secretion impairments may lead to reduced levels of alveolar surfactant. Further studies are in progress to clarify if surfactant synthesis can be enhanced in patients with CDH by prenatal steroids (34) and to ascertain the cause of the reduced amount of surfactant from TA.
The authors are extremely grateful to the nurses of the Neonatal Intensive Care. Without their precious contribution this study would not have been possible. The authors would like to dedicate this study to Marco B. and to his parents, who were extremely supportive to our research efforts. Marco had CDH, he participated in the study, and he is now growing healthy and full of life. The study has received grants from the Department of Medical Research, Veneto Region, Italy and Consorzio Malattie Rare (C.o.M.a.R, Rome, Italy). Received in original form August 7, 2001; accepted in final form March 22, 2002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||