Published ahead of print on October 22, 2004, doi:10.1164/rccm.200406-774OC
© 2005 American Thoracic Society doi: 10.1164/rccm.200406-774OC
Surfactant and Physiologic Responses of Preterm Lambs to Continuous Positive Airway PressureDivision of Pulmonary Biology, Cincinnati Children's Hospital, University of Cincinnati School of Medicine, Cincinnati, Ohio; Fisher & Paykel Healthcare, Ltd., Auckland, New Zealand; and School of Women's and Infant's Health, University of Western Australia, Perth, Australia Correspondence and requests for reprints should be addressed to Alan H. Jobe, M.D., Ph.D., Cincinnati Children's Hospital, Division of Pulmonary Biology, 3333 Burnet Avenue, Cincinnati, Ohio 45229-3039. E-mail: alan.jobe{at}cchmc.org
Although continuous positive airway pressure (CPAP) is used frequently for preterm infants, the relationships between the amount of surfactant and lung physiologic and injury responses to CPAP are unknown. Therefore, saturated phosphatidylcholine (Sat PC) was measured to quantify the surfactant necessary for preterm lambs to breathe successfully on a CPAP of 5 cm H2O (CPAP 5). Five of 21 lambs delivered at 130–136 days gestation failed to keep PCO2 below 100 mm Hg by 2 hours. The lambs that failed had less than 1.9 µmol/kg Sat PC in bronchoalveolar fluid (approximately 3% the pool size at term), less surfactant secretion, and less large aggregate surfactant. Physiologic responses of other 132-day preterm lambs after 2 or 6 hours of CPAP 5, 8 cm H2O CPAP (CPAP 8), or mechanical ventilation were then characterized. At 6 hours, oxygenation and lung gas volumes were higher with CPAP 8 relative to the other groups and  E was decreased with CPAP 8 relative to CPAP 5. Lung dry/wet ratios were greater for the CPAP groups than for the mechanical ventilation group. A small amount of endogenous Sat PC is required for preterm lambs to breathe successfully with CPAP. CPAP 8 improves early newborn respiratory transition relative to CPAP 5.
Key Words: cytokines • lung injury • phosphatidylcholine • respiratory distress syndrome Continuous positive airway pressure (CPAP) was developed for infants with respiratory distress syndrome (RDS) by Gregory and associates in 1971 (1). Although the use of CPAP was associated with lower neonatal deaths from RDS, mechanical ventilation (MV) and surfactant replacement therapy displaced CPAP as primary treatments for RDS. Recently, CPAP is again being used as a primary therapy for neonatal RDS (2). Potential benefits of CPAP may include a reduction in ventilator-associated lung injury and the airway trauma associated with endotracheal intubation (3, 4). RDS is a disease of decreased lung compliance due to inadequate surfactant and increased chest wall compliance secondary to prematurity. Lung atelectasis and edema result in low FRC, hypoxia, hypercarbia, acidosis, pulmonary hypertension, increased work of breathing, and eventual respiratory failure. The use of CPAP to treat RDS decreases atelectasis and allows a larger FRC to be maintained (5), helps stabilize the compliant chest wall of the preterm infant, and decreases thoracoabdominal asynchrony and the work of breathing (6). The improved lung mechanics facilitates improved gas exchange, allows lower respiratory rates, and decreases the incidence of apnea of prematurity (7–10). Although these physiologic and gas exchange benefits have been demonstrated in humans (5–10), there is no quantitative information about the relationship between the amount of surfactant present in a preterm lung and the responses of that lung to CPAP. Because such measurements are not possible in the preterm human, the relationship between endogenous surfactant pools and the physiologic responses to CPAP were evaluated in preterm lambs. The study was then extended to assess if different levels of CPAP were preferable to MV. Some of the results of these studies were previously reported in abstract form (11, 12).
Prenatal Treatments Cincinnati Children's Hospital and the Western Australian Department of Agriculture approved all animal use procedures. Pregnant Australian ewes with twin fetuses were treated with epostane (20 mg intraveneously) and betamethasone (25 mg intramuscularly) 40 hours before preterm Cesarean delivery at 130–136 days gestational age (term = 150 days) to facilitate spontaneous breathing after delivery (3).
Delivery and Ventilation
Pressure–Volume Curves and Lung Processing Mucosa from the distal trachea also was snap frozen. Saturated phosphatidylcholine (Sat PC) was isolated using osmium tetroxide and quantified by phosphorus assay (16). Large aggregate surfactant forms were isolated from fresh BALF from lambs to measure the percent of Sat PC in large aggregates relative to the total Sat PC pool (17).
Lung Injury Markers
Statistics
Surfactant Sat PC Pools and CPAP Twenty-one preterm lambs at 130–136 days gestational age were placed on a CPAP 5 for 2 hours and PCO2 values were used to evaluate respiratory status. A lamb with clear respiratory insufficiency that was dying or was unable to regulate the PCO2 to less than 100 mm Hg at 2 hours of age was considered a CPAP failure. By this criterion, 5 of 21 lambs failed on CPAP 5 (3 of which did not survive to 2 hours). The lambs that failed CPAP were approximately 3 days younger than the 16 lambs that succeeded with CPAP (Table 1). Their respiratory failure was characterized by an inability to achieve adequate VT values despite adequate oxygenation. Pressure–volume curves demonstrated significantly lower lung gas volumes.
To assess if the alveolar surfactant pool size correlated with the respiratory failure, BALF Sat PC pool sizes were compared with final Pco2 values for lambs stratified into three gestational age groups, 130 to 131 days, 132 to 133 days, and 134 to 136 days (Figure 1). Five of six lambs with less than 1.9 µmol/kg Sat PC had PCO2 values greater that 100 mm Hg. The lambs with a Sat PC pool size greater than 1.9 µmol/kg successfully maintained a mean PCO2 of 59 ± 2 mm Hg at 2 hours of age. Interestingly, the PCO2 values were similar for all lambs that succeeded on CPAP 5 despite a wide range in BALF Sat PC up to 18 µmol/kg. BALF Sat PC pool sizes tended to increase with gestational age, but there was considerable overlap between the gestational age groups. In comparing the characteristics of the Sat PC between lambs that failed with those who succeeded on CPAP 5, the mean BALF Sat PC pool sizes were approximately six times larger in the successful group (Figure 2B). A trend toward a larger total lung Sat PC pool size (BALF + lung homogenate) was also noted (p = 0.07) (Figure 2A). The percent of the total lung Sat PC secreted was more than three times higher for the successful lambs (Figure 2C). Furthermore, large aggregate surfactant, which is the surface active form of the surfactant, also was higher for the lambs that succeeded on CPAP 5 (Figure 2D).
Comparison of CPAP 5, CPAP 8, and MV Most of the lambs that were older than 130 to 131 days gestation maintained their PCO2 values for 2 hours. Therefore, lambs at 131 to 133 days gestation were used to evaluate two CPAP levels in comparison to MV for either 2 or 6 hours.
The lung function measurements were focused on the lambs studied for 6 hours because respiratory status was much more consistent after the initial 2 hours of respiratory adaptation. In the CPAP 5 group, 12 of 19 lambs survived for 6 hours, and in the CPAP 8 group, 12 of 19 lambs also survived for 6 hours. One of 11 mechanically ventilated lambs died of a pneumothorax while being ventilated with a high peak inspiratory pressure of 34 cm H2O. BALF Sat PC pool sizes from eight lambs that failed CPAP were 1.1 ± 0.4 µmol/kg, and the percent secretion was 2.2 ± 0.7%, similar to values for the lambs that did not successfully regulate PCO2 at 2 hours in the initial protocol (Figures 2A and 2C). The lambs in the CPAP 5, CPAP 8, and MV groups that survived to 6 hours had similar gestational ages, birth weights, and pH values (Table 2). The oxygenation index was significantly lower at 6 hours for the CPAP 8 lambs than for the mechanically ventilated group. The PO2/FIO2 ratio, an indicator of oxygenation efficiency, was also better for the CPAP 8 lambs than for the CPAP 5 or mechanically ventilated lambs from 2 to 6 hours of age (Figure 3A). CPAP 5 and CPAP 8 lambs had higher PCO2 values than the mechanically ventilated lambs until 2 hours, after which all lambs had similar PCO2 values (Figure 3B). The CPAP 8 and mechanically ventilated lambs achieved similar PCO2 values with similar respiration rates and VT values, with
Surfactant Sat PC Pools at 2 and 6 Hours Sat PC pool sizes were measured for the lambs studied for 6 hours, for lambs randomized to CPAP 5, CPAP 8, or MV for 2 hours, and for seven unventilated control lambs (gestational age 132 days and birth weight 2.8 ± 0.2 kg). For the 2-hour study, 8 lambs were randomized to CPAP 5, 10 to CPAP 8 and 8 to MV. These animals had the same gestational ages and birth weights as the 6-hour group. The total lung and BALF Sat PC pool sizes were similar for the CPAP 5 and CPAP 8 lambs at 2 and 6 hours, but the percent secretion was increased significantly at 2 hours in comparison to 6 hours (Figure 5). The mechanically ventilated lambs had lower total lung Sat PC pool sizes at 2 hours, probably because only one lamb failed MV and thus there was no selection for lambs with larger pool sizes for the MV group.
Indicators of Lung Injury Protein levels, cell numbers and H2O2 production in the BALF were measured at 6 hours of age. The values were low and not different between the groups (Table 2). Monocytes were the major cells in the BALF with no mature macrophages and few granulocytes in the BALF. In comparison to lung tissue of unventilated lambs, IL-1ß mRNA was increased five- to sevenfold in all groups at 2 hours (Figure 6). IL-6 mRNA also was increased for the MV group at 2 hours but no increases in IL-8 were detected. No consistent increase in cytokine mRNA from the lung parenchyma was detected at 6 hours. In contrast to the lung parenchyma, at 6 hours the mucosa of the distal trachea of the CPAP and mechanically ventilated lambs had increased levels of IL-1ß and IL-8 mRNA relative to unventilated lambs (Figure 7).
The best approach to the initial management of preterm infants at risk of RDS has not been defined by clinical trials. The debated variables are which infants should be intubated and electively treated with surfactant and, if CPAP is used initially, what defines CPAP failure and what is the best pressure level of CPAP to use (21). This study demonstrated in the preterm lamb that CPAP failure was caused primarily by a low surfactant pool size of less than approximately 1.9 µmol/kg or 3 mg/kg. This estimate assumes that the surfactant contains approximately 50% Sat PC by weight. The lambs that failed CPAP 5 had less of their total lung Sat PC secreted, and less of the BALF surfactant was in a functional large aggregate form than those who succeeded. We anticipated variable degrees of respiratory distress with different PCO2 values. However, the lambs had either severe respiratory failure or the ability to regulate PCO2 values if the surfactant pool was greater than approximately 3 mg/kg. In this lamb model, high PCO2 values that do not progressively decrease indicated the need for an intervention (surfactant and/or MV). Our results are consistent with the clinical observations that some very preterm infants with mild RDS can breathe successfully with CPAP alone despite surfactant pool sizes that may be small. Our results also are consistent with the clinical observations that some infants at high risk of RDS can be intubated, treated with surfactant, and extubated to CPAP as the major problem is surfactant deficiency (22). However, surfactant treatments were not attempted in this protocol.
After establishing the model of CPAP in lambs delivered by Cesarean section, CPAP levels of 5 or 8 cm H2O were evaluated. Five cm H2O pressure were used because it is commonly used clinically. If surfactant deficiency is the primary problem, a higher level of CPAP should be helpful in improving oxygenation and better recruiting and sustaining the FRC (6). Because CPAP was initiated without any MV after Cesarean section, the lungs contained the fetal lung fluid that remained despite the betamethasone and epostane treatments and that did not drain passively at delivery. Therefore, all lambs were given CPAP 8 for the first 10 minutes. The same proportion of lambs failed CPAP 5 and CPAP 8 during the 6 hours of study, suggesting that CPAP 8 did not allow more lambs to tolerate CPAP. However, the lambs on CPAP 8 had better oxygenation and more efficient breathing as indicated by the lower The surfactant pool size of less than 3 mg/kg that resulted in lethal RDS without MV was similar to the mean value of approximately 5 mg/kg reported by Adams in 1970 for infants who died of RDS without MV (26). The standard dose for the treatment of RDS is 100 or 200 mg/kg, depending on the surfactant product used. To put these values into context, the term sheep or rabbit has an alveolar pool size of approximately 100 mg/kg surfactant (27, 28). Although no measurements are available for the term human, large amounts of surfactant can be recovered from amniotic fluid, indicating that large pool sizes are also present in the term human (29). The adult human has an alveolar pool size of approximately 5 mg/kg, whereas the value for the sheep is approximately 20 mg/kg (30, 31). Surfactant function is dependent on both the pool size and the quality of the surfactant. Surfactant from the preterm has decreased surface-active function and is more susceptible to inactivation by products of lung inflammation/injury than surfactant from mature animals (17). Although surfactant pool size for the preterm lamb with lethal RDS is not much less than for the adult human, the surfactant may be less effective and more dilute because of increased amounts of alveolar fluid. An incidental observation was that the percent of secreted Sat PC was increased at 2 hours relative to 6 hours with both CPAP 5 and CPAP 8. A possible explanation is that secretory stimuli should be maximal after preterm birth, whereas activation of recycling and catabolic pathways may be delayed. Rapid changes in surfactant pools just before and after birth have been reported previously (32, 33). An advantage of using CPAP to help with the spontaneous initiation of breathing after birth is the fact that the preterm lung is not injured initially. If injury can be minimized, then less surfactant should be needed because inactivation will not occur (34). Endogenous surfactant is much more effective in small amounts than exogenous surfactant. In preterm ventilated rabbits, compliance improves for small increases in endogenous surfactant, whereas much more surfactant given as a treatment is required to achieve smaller increases in compliance (35). Said another way, the dose–response curve for endogenous surfactant is much better than for treatment doses of surfactant. This result is explained, in part, by the better distribution of the endogenous than of the exogenous surfactant. The lambs were exposed to antenatal betamethasone as is the standard of care for infants delivering prematurely (36). The antenatal glucocorticoid exposure will increase lung gas volume without increasing surfactant lipid over 40 hours (37). However, the fetal lung may have a better dose–response curve after exposure to glucocorticoids (38, 39). The minimal amount of surfactant required to support CPAP may be less after antenatal glucocorticoid therapy. In a previous study, we found that preterm lambs mechanically ventilated to a PCO2 of approximately 40 mm Hg had increased indicators of injury relative to lambs with PCO2 values of approximately 60 mm Hg on CPAP (3). In this study, the Pco2 values of the mechanically ventilated lambs were successfully matched to those achieved by the CPAP lambs. Gentle MV was utilized to minimize lung injury (40), and no indicators of severe injury were found after 6 hours of CPAP or MV. The proinflammatory cytokine mRNA for IL-1ß was elevated in all groups at 2 hours and not at 6 hours. This result differs from our previous experiences with MV in preterm lambs, probably because permissive hypercapnea was accepted (13, 41). The IL-6 mRNA level in lung parenchyma was the only measurement that distinguished the CPAP groups from the mechanically ventilated lambs. A caveat for these experiments is that the ewes were given betamethasone approximately 40 hours before studying the lambs because during previous experiments preterm lambs did not breathe after Cesarean delivery (3). Betamethasone treatment may blunt an inflammatory response in the preterm lamb lung (42). Proinflammatory cytokine expression was measured in the mucosa from the distal trachea because the airways are a major site of injury in the preterm lung (43). Although the IL-1ß or IL-8 mRNA increases could not be detected in lung parenchyma at 6 hours, these proinflammatory cytokine mRNAs were increased in the tracheal mucosa from the endotracheal CPAP and the mechanically ventilated lambs at 6 hours. The airway responses may result from the endotracheal tube, although the mucosa was sampled distal to the tip of the tube. Heated and humidified respiratory gases were used from birth for these experiments to minimize airway injury. Future experiments will be required to learn if the airway mucosal responses of the preterm lung will occur with nasal CPAP.
The authors thank Sanofi-Synthelabo, Malvene, PA, for the gift of epostane.
Supported by grant HD-12714 from the National Institutes of Child Health and Development and Fisher & Paykel, Auckland, New Zealand. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Conflict of Interest Statement: N.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; Z.C. is an employee of Fisher & Paykel Healthcare and is employed as a full-time Clinical Researcher and entitled to renumeration and benefits similar to those employees of Fisher & Paykel Healthcare in a similar position with similar responsibilities; T.J.M.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; I.N. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.I. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; A.H.J. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form June 18, 2004; accepted in final form October 19, 2004
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
p < 0.01 vs. CPAP 8). (D) The CPAP 5 group had elevated 
