on Pulmonary Gas Exchange and Mechanics in Premature Lambs With Respiratory Distress Syndrome
a Research Unit for Experimental Neonatal Respiratory Physiology, Department of Pediatrics, Hospital Cruces, University of the Basque Country, Barakaldo, Bizkaia, Spain
b Neonatal Intensive Care Unit, Department of Pediatrics, Hospital Cruces, University of the Basque Country, Barakaldo, Bizkaia, Spain
| ABSTRACT |
|---|
|
|
|---|
OBJECTIVE. The objective was to compare the acute and sustained effects of lucinactant among premature lambs with respiratory distress syndrome (RDS) with the effects of a natural porcine surfactant (poractant-
).
METHODS. After 5 minutes of mechanical ventilation twin premature lambs were assigned randomly to the lucinactant group (30 mg/mL, 5.8 mL/kg) or the poractant-
group (80 mg/mL, 2.2 mL/kg). Heart rate, systemic arterial pressure, arterial pH, blood gas values, and lung mechanics were recorded for 12 hours.
RESULTS. Baseline fetal pH values were similar for the 2 groups (pH 7.27). After 5 minutes of mechanical ventilation, severe RDS developed (pH: <7.08; PaCO2: >80 mm Hg; PaO2: <40 mm Hg; dynamic compliance: <0.08 mL/cm H2O per kg). After surfactant instillation, similar improvements in gas exchange and lung mechanics were observed for the lucinactant and poractant-
groups at 1 hour (pH: 7.3 ± 0.1 vs 7.4 ± 0.1; PaCO2: 8 ± 18 mm Hg vs 40 ± 8 mm Hg; PaO2: 167 ± 52 mm Hg vs 259 ± 51 mm Hg; dynamic compliance: 0.3 ± 0.1 mL/cm H2O per kg vs 0.3 ± 0.1 mL/cm H2O per kg). The improvements in lung function were sustained, with no differences between groups. Cardiovascular profiles remained stable in both groups.
CONCLUSIONS. Among preterm lambs with severe RDS, lucinactant produced improvements in gas exchange and lung mechanics similar to those observed with a porcine-derived surfactant.
Key Words: natural surfactant synthetic surfactant respiratory distress syndrome premature lambs
Abbreviations: FIO2fraction of inspired oxygen MAPmean airway pressure OIoxygenation index PEEPpositive end-expiratory pressure PIPpeak inspiratory pressure RDSrespiratory distress syndrome SPsurfactant protein
In 1959, Avery and Mead1 reported that deficiency of pulmonary surfactant had a key role in the pathogenesis of neonatal respiratory distress syndrome (RDS), and it is now accepted that RDS is primarily a transient surfactant deficiency state. Mechanical ventilation contributes significantly to the treatment of premature infants with RDS, but it was not until exogenous surfactant therapy was introduced by Fujiwara et al2 in 1980 that the outcomes of preterm infants with RDS improved significantly.
Synthetic surfactants36 and natural surfactants derived from bovine lungs,7,8 porcine lungs,79 and human amniotic fluid10 were shown to be effective in the prevention and treatment of neonatal RDS. Those studies showed not only acute improvement in pulmonary gas levels but also an increase in lung compliance.
Natural surfactants derived from animal lungs, containing both phospholipids and associated surfactant proteins (SPs), were shown to be effective in different experimental models of RDS.1114 Similarly, animal-derived surfactants were shown to be effective among infants with RDS,810 decreasing rates of death and morbidity associated with RDS.
Different protein-free, synthetic surfactant preparations have been tested in clinical trials among neonates with RDS36 and in experimental models.1517 Clinical trials of natural versus synthetic surfactants showed controversial results but, in general, animal-derived surfactants have been found to be more efficacious.15,16,1821
Surfactant-associated proteins, present in animal-derived surfactants, are important in the development of surfactant function. More specifically, SP-B seems to be essential in maintaining alveolar expansion and reduces surface tension in combination with phospholipids.22,23 A second generation of synthetic surfactants with synthetic analogs of SPs have been developed and tested in experimental models.2427 In 1991, Cochrane and Revak28 described a new synthetic surfactant, lucinactant, containing the phospholipids dipalmitoylphosphatidylcholine and palmitoyloleoylphosphatidylglycerol, palmitic acid, and a 21-residue synthetic peptide, called KL4, that mimics the native sequence of SP-B and thus emulates its surface-active enhancement action, increasing the surfactant activity of the phospholipids.
Lucinactant has been shown to be effective in animal models of acute RDS29 and meconium aspiration syndrome30 and among human infants with RDS.31,32 It has been also tested in uncontrolled trials of acute RDS33 and neonatal RDS34 and among term infants with meconium aspiration syndrome.35 Although its effects are being tested, to date its efficacy has not been compared with that of a natural surfactant preparation considered to be the standard by most neonatologists.
The aim of this study was to compare the acute and sustained effects of lucinactant with those of a natural surfactant preparation in an experimental model of RDS in premature lambs. We hypothesized that the effects of the synthetic surfactant on pulmonary gas exchange and mechanics would be similar to those obtained with the natural surfactant poractant-
.
| METHODS |
|---|
|
|
|---|
Animal Preparation
Ewes were sedated with xylazine (68 mg, administered intramuscularly) and ketamine (5 mg/kg, administered intravenously), and anesthesia was maintained with propofol (3040 mg/kg per hour); Ringer's lactate solution was infused as needed. A tracheal tube (inner diameter: 8 mm) was inserted into the trachea and connected to a volume-controlled ventilator (CPU-1; Ohmeda, Riverside, CA), to maintain adequate gas exchange (initial settings as follows: rate: 30 cycles per minute; peak inspiratory pressure [PIP]/positive end-expiratory pressure [PEEP]: 1416/2 cm H2O; inspiratory/expiratory ratio: 1:2; fraction of inspired oxygen [FIO2]: 0.40.6). A peripheral arterial cannula (Insyte; BD, Madrid, Spain) was inserted to monitor arterial pressure and blood gas levels. With the ewe lying on its right side, the uterus was exposed with a lateral subcostal incision and the head and neck of the fetal lambs were exteriorized. A rubber glove was placed over the snout to prevent spontaneous breathing; however, because of the anesthesia, breaths were rarely observed. A tracheal tube (Hi-Lo Jet tracheal tube; inner diameter: 4.0 mm; Mallinckrodt Medical, St Louis, MO) was inserted through a tracheotomy, with its tip located above the carina, and was tied around the trachea to prevent leaks. During these procedures, lungs were allowed to drain slowly through gravity. End-hole catheters (8-French XRO umbilical catheters; Vygon, Ecouen, France) were inserted into the jugular vein and carotid artery. Lambs were given ketamine (8 mg) and pancuronium bromide (0.4 mg), administered intravenously, and the umbilical cord was cut.
Lambs were weighed (mean body weight: 3.5 ± 0.4 kg), dried, and placed in a servo-controlled radiant warmer to maintain the rectal temperature at 38°C to 39°C. Then the tracheal tube was connected to a time-cycled, pressure-limited ventilator (Bourns BP-200; Beard Medical Systems, Riverside, CA), with the following initial settings: rate: 40 cycles per minute; PIP/PEEP: 25/5 cm H2O; inspiratory/expiratory ratio: 1:2; FIO2: 1.0; flow rate: 10 L/minute. A constant FIO2 of 1.0 was maintained throughout the experiment, with the other initial parameters being changed to maintain adequate arterial blood gas levels (maximal PIP of 35 cm H2O and rate of 60 cycles per minute). Ketamine was infused in 5% glucose (4 mg/kg per hour). A mean arterial pressure of 40 mm Hg was maintained through infusion of 10 mL/kg heparinized ewe's blood, if necessary.
Experimental Design
Premature lambs were assigned randomly to receive poractant-
(Curosurf; Chiesi Farmateutici, Parma, Italy) or lucinactant (Surfaxin; Discovery Laboratories, Doylestown, PA). In cases of twins, each received a different surfactant, on the basis of the order of randomization. The 2 experimental groups were managed equally, as follows.
Lucinactant Group
Lambs received lucinactant at 175 mg/kg (30 mg/mL). Surfactant was handled according to the manufacturer's instructions. Vials were heated at exactly 44°C for 15 minutes, with a heater block (Isotemp dry bath model 147; Fisher Scientific, Pittsburgh, PA), and were shaken vigorously to obtain a uniform suspension. One half of the total dose (5.8 mL/kg) was drawn into separate syringes (with an extra volume of 0.5 mL of air). The first one half of the total dose was delivered with the lamb in the right lateral decubitus position, as a bolus infused in 3 to 5 seconds through the lateral side-port of the tracheal tube, without disconnecting the animal from the ventilator and maintaining PEEP during administration of the drug. The lambs were hand-ventilated for 30 seconds at a rate of 60 cycles per minute, with a PIP adequate for good chest wall excursion (not exceeding 30 cm H2O). The second half-dose was administered exactly the same way, with the lamb in the left lateral decubitus position.
Poractant-
Group
Animals received poractant-
at 175 mg/kg (80 mg/mL), instilled as noted on its label. Vials were warmed by hand for 2 to 3 minutes and swirled gently to mix the contents. A total dose of 2.2 mL/kg was prepared as 2 half-doses in separate syringes (with an extra volume of 0.5 mL of air). Otherwise, lambs were treated as described for the previous group.
Measurements
Heart rate and arterial blood pressure (systolic, diastolic, and mean) were measured and recorded continuously (OmniCare, GMS 24; Hewlett Packard, Boeblingen, Germany). Arterial pH, PaO2, PaCO2, and base excess (AVL model 945; AVL Scientific Corp, Schaffhausen, Switzerland), as well as lung mechanics (tidal volume, dynamic compliance, resistance, inspiratory/expiratory ratio, and frequency), were measured 5, 15, and 30 minutes after surfactant instillation and then every 30 minutes until the end of experiment, at 12 hours. The oxygenation index (OI) was calculated as [mean airway pressure (MAP) (cm H2O) x FIO2/PaO2 (mm Hg x 100)].34 To assess pulmonary ventilation independent of ventilator settings, the ventilatory efficiency index was calculated as [3800/(PIP PEEP) (cm H2O) x respiratory rate (breaths per minute) x PaCO2].38
Pulmonary Mechanics
Lung dynamic compliance, airway resistance, tidal volume, and minute ventilation were calculated with a computerized system (Peds, Medical Associated Services laboratory version; PTI, Hatfield, PA). To evaluate lung overdistention, the compliance for the terminal 20% of the inspiration was calculated in relation to the compliance for the completed breath.39
Airflow was measured during the entire respiratory cycle with a pneumotachometer (Fleish 00; OEM Medical, Richmond, VA), and airway pressure was measured with a differential pressure transducer (Validyne model MP45; Engineering Corp, Northridge, CA). Simultaneous signals for airflow and transpulmonary pressure (difference between intrapleural and airway pressures) were recorded and used to calculate pulmonary mechanics with least-mean squares analysis.40 Data for each breath were reviewed automatically to meet all selection criteria, before being included in the averaged data. Ten random breaths were analyzed to represent pulmonary function for each lamb at each time point. At the end of the experiment, lambs were killed with an overdose of potassium chloride
Statistical Analyses
Values are expressed as mean ± SEM. The sample size was calculated with G-Power (version 2.0; Department of Psychology, University of Bonn, Bonn, Germany), and at least 216 determinations were necessary to obtain a power of 0.95 with an
error of .05 for the study. Comparisons of measured values before and after surfactant treatment were assessed with single-factor analysis of variance (Statview; Abacus Concepts, Berkeley, CA). Comparisons between groups were tested with 2-factor analysis of variance for repeated measures as a function of time and group. Both analyses were corrected with the Bonferroni-Dunn correction. P values of <.05 were accepted as significant.
| RESULTS |
|---|
|
|
|---|
|
group responded to surfactant treatment. In the lucinactant group, 2 lambs did not show an increase in PaO2, although PaCO2 decreased, after surfactant instillation; data for these nonresponding animals were included in statistical analyses, graphs, and tables.
Survival rates were similar; 1 animal in each group did not reach the end of the experiment, dying at 10 and 10.5 hours in the lucinactant and poractant-
groups, respectively. These 2 lambs died after a cardiac arrest, after a period of extremely low systemic arterial blood pressure that was unresponsive to infusions of both volume (dam's blood) and dopamine (up to 10 µg/kg per minute).
Pulmonary Gas Exchange
Fetal PaO2 values were similar in the lucinactant and poractant-
groups (31.7 ± 0.8 mm Hg vs 30.1 ± 3.9 mm Hg), as was basal PaO2 after 5 minutes of mechanical ventilation (41.4 ± 4.7 mm Hg vs 33.9 ± 9.2 mm Hg). In both groups, there were significant increases in PaO2 recorded at 5 and 15 minutes after tracheal surfactant instillation. PaO2 values remained above 250 mm Hg for the entire experiment (Fig 1A). For the 2 animals that did not respond to lucinactant, PaO2 values remained below 100 mm Hg, at levels similar to those obtained 5 minutes after the start of mechanical ventilation, before surfactant administration.
|
administration and from 30.4 ± 3.7 to 19.5 ± 8.1 after lucinactant instillation. Values in both groups remained stable and similar to those reached after 60 minutes of surfactant delivery. The decrease in OI values was faster and more pronounced in the poractant-
group (P < .05) (Table 2).
|
: 54.9 ± 2.6 mm Hg). Similarly, no differences were observed between groups in basal values of PaCO2 after the first 5 minutes of mechanical ventilation (lucinactant: 81.1 ± 16.1 mm Hg; poractant-
: 90.8 ± 10.7 mm Hg). After surfactant instillation, PaCO2 levels did not decrease significantly until 60 minutes (lucinactant: 47.9 ± 7.6 mm Hg; poractant-
: 40.4 ± 7.7 mm Hg), remaining in the normal range until the end of experiment, without differences between groups.
Initial ventilatory efficiency index values were also similar (lucinactant: 0.03 ± 0.01; poractant-
: 0.04 ± 0.01), increasing 60 minutes after surfactant instillation to 0.09 ± 0.03 and 0.12 ± 0.03 in the lucinactant and poractant-
groups, respectively. Values continued to increase until 4 hours (lucinactant: 0.32 ± 0.08; poractant-
: 0.30 ± 0.08). Thereafter, the 2 groups maintained comparable values until the end of the experiment (Table 3).
|
7.35 thereafter (Table 3).
Ventilator Settings
All animals underwent ventilation throughout the study, with a FIO2 of 1.0. Mean basal ventilator settings were similar for the 2 groups, without statistical differences in rate (lucinactant: 58.3 ± 1.6 cycles per minute; poractant-
: 58.3 ± 1.6 cycles per minute), PIP (lucinactant: 26.7 ± 1 cm H2O; poractant-
: 26.7 ± 1 cm H2O), PEEP (lucinactant: 5 ± 0 cm H2O; poractant-
: 5 ± 0 cm H2O), MAP (lucinactant: 12.2 ± 0.3 cm H2O; poractant-
: 12.2 ± 0.3 cm H2O), inspiratory time (lucinactant: 0.3 ± 0 seconds; poractant-
: 0.3 ± 0 seconds), or expiratory time (lucinactant: 0.7 ± 0 seconds; poractant-
: 0.7 ± 0 seconds).
After instillation of both surfactant preparations, ventilator settings were changed to provide adequate chest wall excursion. At 30 minutes of treatment, there were statistical differences in PIP (lucinactant: 31.2 ± 0.8 cm H2O; poractant-
: 27.5 ± 1.1 cm H2O) and MAP (lucinactant: 14.1 ± 0.5 cm H2O; poractant-
: 12.5 ± 0.3 cm H2O). Ventilator adjustments were different in the 2 groups, because greater PIP, MAP (Fig 1B), inspiratory time, and expiratory time were required for the lucinactant-treated lambs, to maintained adequate gas exchange (P < .05).
Cardiovascular Profile
Mean fetal heart rates were similar for the 2 groups (lucinactant: 136 ± 17 beats per minute; poractant-
: 168 ± 13 beats per minute). After delivery and surfactant administration, there was an upward trend that was maintained throughout the experiment, without differences between groups (Table 3). For all fetuses, mean systemic arterial pressures were similar (lucinactant: 55 ± 2 mm Hg; poractant-
: 65 ± 4 mm Hg) and increased at 5 minutes of postnatal life (lucinactant: 70 ± 5 mm Hg; poractant-
: 76 ± 5 mm Hg). Although there was a slight decrease in systemic arterial pressure after lucinactant administration, pressure values returned to basal levels within 5 minutes and then were maintained throughout the experiment (Table 3).
Pulmonary Mechanics
Initial mean dynamic compliance was found to be extremely low in both groups (lucinactant: 0.08 ± 0.02 mL/cm H2O per kg; poractant-
: 0.07 ± 0.01 mL/cm H2O per kg). After surfactant treatment, there were similar increases (P < .05) in the 2 groups (lucinactant: 0.21 ± 0.04 mL/cm H2O per kg; poractant-
: 0.22 ± 0.02 mL/cm H2O per kg). By 30 minutes of age, values were significantly higher basal values. Both groups registered an upward trend during the first 4 hours of the experiment and then maintained values until 12 hours (Fig 2).
|
: 1.9 ± 0.6 mL/kg) and increased significantly after surfactant instillation (lucinactant: 5.2 ± 1.2 mL/kg; poractant-
: 4.9 ± 0.6 mL/kg). The tidal volume increased up to 6 hours and maintained values of >10 mL/kg until the end of experiment. The compliance for the terminal 20% of the inspiration/compliance for completed breath ratio did not indicate overdistention in either group; values were >0.8 through the experiment (Table 3). After delivery, airway resistance was high in all animals (lucinactant: 108 ± 15 cm H2O/L per second per kg; poractant-
: 140 ± 31 cm H2O/L per second per kg). In both groups, there were significant decreases 60 minutes after surfactant delivery (lucinactant: 62 ± 6 cm H2O/L per second per kg; poractant-
: 52 ± 5 cm H2O/L per second per kg), and values remained at
50 cm H2O/L per second per kg thereafter. | DISCUSSION |
|---|
|
|
|---|
). The effects of lucinactant on lung function are in agreement with those previously reported in animal models of acute RDS29,31,32 and meconium aspiration,30 as well as for human infants with severe RDS.41,42 The best experimental model to compare the effects of different surfactant preparations probably is a model of lung immaturity related to prematurity. For mixed-Western breed fetal lambs, the gestational age of 135 days might not seem appropriate, because of the variable degrees of lung maturity present at this gestational age.43 However, differences in timing of fetal lung maturation for different strains of sheep are well known. Wolfson and Shaffer44 described a positive correlation between lung compliance and gestational age. With extrapolation of those values to our model and the basal values of in vivo lung compliance measured in the strain used in this study (Basque Latxa strain), the estimated gestational age was 100 ± 6 days.
The overall time courses of all lung function parameters were somewhat different in the 2 groups for the whole 12 hours of the experiment. Although the PaO2 and PaCO2 responses to both surfactants were similar, OI was statistically higher in lucinactant group because a higher MAP was needed to maintain pulmonary gas exchange.
We also observed some differences in the acute responses to dosing. Lucinactant-treated lambs had a slower PaO2 increase (Table 2), which should not be considered a negative result, because a rapid PaO2 increase might be related to a quick decrease in pulmonary vascular resistance, which could cause an increase in pulmonary blood flow attributable to a large left-to-right shunt in a patent ductus arteriosus,2 although these observations have not been confirmed among human subjects.45 Unfortunately, pulmonary vascular resistance was not measured in this study.
It should be noted that, although both surfactant preparations were instilled in the same way and all animals were treated identically, those that received poractant-
required less hand-ventilation immediately after dosing, because the drug was observed up to the endotracheal tube less frequently. These differences could be related to the larger dose volume for the lucinactant group and/or less-homogeneous surfactant distribution. In this study, 2 lambs in the lucinactant group did not respond to surfactant treatment, although they were treated similarly. Unfortunately, no morphologic or biochemical studies were performed on lung samples; therefore, a nonhomogeneous distribution of the surfactant could not be ruled out.
To instill both surfactant preparations, we used a lateral side-port on the tracheal tube, on the basis of the observations by Merritt et al,32 who demonstrated that, with PEEP maintained during instillation, a smaller decrease in arterial oxygen saturation was recorded. Nevertheless, a decrease in arterial oxygen saturation of 25% and the presence of surfactant reflux up the tracheal tube were reported.
We used an identical dose of surfactant (175 mg/kg) to treat all animals. This dose is close to 200 mg/kg, as recommended for poractant-
, and also to the KL4 dose used by Revak et al31 for premature rhesus monkeys, with an optimal oxygenation response. Moreover, this dose regimen was also used in the 2 randomized, clinical trials performed with lucinactant.41,42
Animal-derived surfactants are highly effective in the prevention and treatment of RDS among premature human infants.7,9,46,47 Improvements in gas exchange and lung mechanics and significant decreases in morbidity and mortality rates, without concomitant increases in long-term disability rates, have been reported. Furthermore, animal-derived surfactants have been demonstrated to be superior to synthetic products for very immature infants.20 However, natural preparations also have some limitations, related to limited supplies, lack of homogeneity of composition, potential risks for the transmission of animal-related diseases, and development of immunogenic responses, because of the heterologous proteins present in these preparations.
Several synthetic surfactants containing only surface-active phospholipids, such as dipalmitoylphosphatidylcholine and phosphatidylglycerol, and no surfactant-associated proteins have been developed, with controversial results.26 In fact, only Exosurf has undergone extensive clinical trials and is still used in clinical practice.6,48
Native lung surfactant is a complex mixture of phospholipids and surfactant-related proteins49 that reduces surface tension by forming a surface film in the alveolus.50 The hydrophobic SPs (SP-B and SP-C) enhance the surface adsorption of phospholipid molecules at the alveolar interface,49,51 improving pulmonary gas exchange52 and optimizing static and dynamic lung mechanics.26 Therefore, several groups have developed second-generation synthetic surfactants containing not only surface-active phospholipids but also analogs of the surfactant-associated proteins.18,24,25,27
In 1991, Cochrane and Revak28 described a novel synthetic surfactant, lucinactant (KL4), containing phospholipids and a synthetic peptide that mimics SP-B activity. This drug decreases surface tension in vitro and is less inhibited by meconium and by serum components.53,54 Furthermore, lucinactant has been shown to improve pulmonary gas exchange and mechanics in animal models of RDS29 and meconium aspiration,30 although its effects have not been compared with those of a natural surfactant preparation (currently considered the standard).
An uncontrolled study was published,34 and 2 randomized, clinical trials were published recently. Those studies compared lucinactant with other surfactants, ie, a synthetic surfactant (Exosurf)42 and a natural preparation (Curosurf),41 in preventing RDS among premature infants of <32 and 29 weeks of gestation, respectively. In those trials, lucinactant was found to be more effective than Exosurf in reducing RDS-related mortality rates42 and at least as safe and efficacious as a natural surfactant preparation.41 Although additional studies are required to confirm these observations, it seems that lucinactant, a synthetic surfactant containing a peptide that mimics SP-B action, could be an alternative to natural surfactant preparations.
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
We are grateful to Discovery Laboratories (Doylestown, PA) for making available to us the lucinactant samples for the experiments. We also thank Idoia Aparicio and Ricardo Murias for assistance with the pregnant ewes and Rafael Morales for support in performing the experiments.
| FOOTNOTES |
|---|
Address correspondence to Adolf Valls-i-Soler, MD, Neonatal Intensive Care Unit, Department of Pediatrics, Hospital de Cruces, University of the Basque Country, Plaza de Cruces, Barakaldo E-48903, Bizkaia, Spain. E-mail: avalls{at}hcru.osakidetza.net
The authors have indicated they have no financial relationships relevant to this article to disclose.
| REFERENCES |
|---|
|
|
|---|
for treatment of respiratory distress syndrome in neonates born at 2529 weeks' gestation: a randomised trial.
Lancet. 2000;355
:1387
1392[CrossRef][ISI][Medline]This article has been cited by other articles:
![]() |
A. Almlen, G. Stichtenoth, B. Linderholm, M. Haegerstrand-Bjorkman, B. Robertson, J. Johansson, and T. Curstedt Surfactant proteins B and C are both necessary for alveolar stability at end expiration in premature rabbits with respiratory distress syndrome J Appl Physiol, April 1, 2008; 104(4): 1101 - 1108. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. L. Halliday and C. P. Speer Lucinactant versus poractant-alpha in premature lambs. Pediatrics, June 1, 2006; 117(6): 2324 - 2325. [Full Text] [PDF] |
||||
![]() |
A. Valls-i-Soler, E. Gastiasoro-Cuesta, and F. J. Alvarez-Diaz Lucinactant Versus Poractant-{alpha} in Premature Lambs: In Reply Pediatrics, June 1, 2006; 117(6): 2325 - 2325. [Full Text] [PDF] |
||||
Read all P3Rs
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||