Published ahead of print on April 24, 2003, doi:10.1164/rccm.200301-121OC
© 2003 American Thoracic Society
Therapeutic Lung Lavage in the Piglet Model of Meconium Aspiration SyndromeDepartments of Neonatology, Anatomical Pathology, and Radiology, Royal Children's Hospital; and Murdoch Childrens Research Institute, Melbourne, Australia Correspondence and requests for reprints should be addressed to Peter Dargaville, M.D., Department of Neonatology, Royal Children's Hospital, Flemington Road, Parkville, VIC 3052, Australia. E-mail: peter.dargaville{at}rch.org.au
Therapeutic lung lavage is an emerging treatment for meconium aspiration syndrome. Our objective was to investigate the type of fluid and aliquot volume most appropriate for lung lavage in this condition. Meconium injury was induced in 2-week-old piglets, followed by a 30 ml/kg lavage in two aliquots 40 minutes later. Lavage with either dilute bovine surfactant (2.5 mg/ml) or a perfluorocarbon emulsion (20% wt/vol) improved oxygenation compared with a nonlavaged control group, but only with dilute surfactant was there a sustained improvement in oxygenation (alveolar–arterial oxygen difference at 5 hours: dilute surfactant 250 mm Hg; perfluorocarbon emulsion 460 mm Hg; controls 460 mm Hg; p = 0.0031). There was histologic and biochemical evidence of decreased lung injury in the dilute surfactant group. In a further study, 30 ml/kg dilute surfactant lavage was performed 40 minutes after meconium injury using either two aliquots of 15 ml/kg, or multiple 3-ml aliquots. Aliquot volume of 15 ml/kg was associated with increased meconium removal, better post-lavage lung function, and less lung injury. Dilute surfactant lavage using two 15-ml/kg aliquots is an effective therapy in the piglet model of meconium aspiration, and should be evaluated in human infants with this condition.
Key Words: meconium aspiration syndrome • pulmonary surfactants • pulmonary lavage • perfluorocarbon Meconium aspiration syndrome (MAS) is an important cause of respiratory distress in the term newborn infant, and may result in considerable respiratory morbidity (1, 2). The pathophysiology of MAS largely reflects the noxious effects of meconium in the tracheobronchial tree, and includes airway obstruction (3), epithelial damage (4), surfactant inhibition (5), inflammation (2), and pulmonary hypertension (2). These elements combine to produce lung dysfunction that can be extremely difficult to treat, often requiring the use of nonconventional means of respiratory support, including high frequency ventilation and nitric oxide therapy (1, 2). Current therapies for MAS are essentially supportive and, apart from the effect of bolus surfactant therapy on the use of extracorporeal membrane oxygenation (6), appear to do little to alter the natural course of the disease. Recently, however, therapeutic lung lavage has been investigated in MAS, with the intent of slowing or halting the progression of the disease by removing meconium from the lung. In animal models of MAS, saline (7–13), perfluorocarbon (8, 13–17), and full-strength (12) or diluted (9–11, 13, 17–19) surfactant have been investigated as lavage fluids. The total volume of fluid used for lavage in these studies has varied from 5 to 80 ml/kg, with aliquot volume (the amount of fluid instilled into the lung at one time) ranging from as little as 2 ml up to 10 ml/kg. Important findings to date are that lung lavage using surfactant results in better gas exchange and/or pulmonary mechanics than either no lavage (11–13, 17, 19) or bolus surfactant therapy (11), that saline is an inferior lavage fluid to surfactant (9, 10, 12, 13) or perfluorocarbon (8), and that lavage removes considerable amounts of meconium from the lung (9–11, 19). The possibility of exploiting the unique properties of perfluorocarbon as a lavage fluid has been explored in several animal models of MAS (8, 13–17). Although periods of tidal liquid ventilation have been found to improve oxygenation and lung histology, discrete episodic lavage with pure perfluorocarbon does not appear to confer any advantage over lavage with equal volumes of dilute surfactant (13, 17). This may be due to the relatively high density of perfluorocarbon, and/or the immiscibility of biological fluids in perfluorochemicals. Both these factors might be overcome by using an emulsion of perfluorocarbon and aqueous fluid for lavage, but this has not been investigated to date. In human infants, the available data on lavage therapy in MAS are limited to several case series (20–25) and two small randomized controlled trials (26, 27). These studies have used both saline (20, 21, 24) and dilute surfactant (22–27) as lavage fluids, with total lavage volumes of 3 to 48 ml/kg, and aliquot volumes ranging from 2 to 15 ml/kg. Considerable amounts of meconium appear to be removed by lavage, with improvement in postlavage lung function noted in studies comparing a lavage group with either historical (22, 24) or randomized (27) controls in whom lavage was not performed. The diverse fluid types, lavage volumes, and aliquot volumes used in the studies cited above bespeak the need for a systematic examination of the factors that may affect both the safety and efficacy of lung lavage in MAS. In this experimental series, we have investigated some of these factors in the piglet model of MAS. We sought to identify an optimal total lavage volume, and then in separate randomized trials, to compare different lavage fluids and aliquot volumes. We hypothesized that meconium recovery and postlavage lung function would be influenced both by the type of lavage fluid and by aliquot volume. Some of the results of the studies in this paper have previously been reported in abstract form (28, 29).
Animal Preparation The animal experimentation procedures in this study complied with the guidelines of the Australian National Health and Medical Research Council, and were approved by our institutional Animal Ethics Committee. Two-week-old Landrace–Lancaster cross piglets were anesthetized, intubated, and ventilated as described in the online supplement. Venous and arterial cannulae were sited and, in seven preliminary animals, pulmonary artery catheters were inserted for continuous monitoring of pulmonary artery pressure (Ppa). Initial ventilator settings were peak inspiratory pressure (PIP) 15–20 cm H2O, positive end-expiratory pressure (PEEP) 4 cm H2O, respiratory rate 20 breaths/minute, inspiratory time 0.6 seconds, gas flow 15 L/minute, and FIO2 1.0. In phases II and III of the study, VT and compliance and resistance of the respiratory system (Crs, Rrs) were displayed in real time (VenTrak respiratory monitoring equipment; Novametrix Medical Systems Inc., Wallingford, CT). VT values were a mean of the previous 60 seconds recording, whereas Crs, and Rrs were manually averaged over 10 breaths. Flow and pressure calibration of the equipment was checked daily against a reference source (RT200 Calibration Analyzer; Timeter, Lancaster, PA).
Induction of Meconium Injury Once the anesthetized animal was stable on the ventilator, meconium injury was induced as outlined in the online supplement. In Phase I of the study, 4 ml/kg of 20% meconium was instilled, followed immediately by lung lavage as described below. In Phases II and III, the desired end-point was arterial oxygen saturation (SaO2) < 92%; if, 5 minutes after instillation of 4 ml/kg of meconium, this target was not achieved, a further 1–2 ml/kg of 20% meconium was administered.
Phase I Determination of Optimal Total Lavage Volume Once the lavage was completed, phase I animals received further saline lavage as part of another research study. The amount of meconium recovered during further lung lavage was negligible (data not shown).
Phase II Controlled Trial of Different Lavage Fluids The experimental model of MAS was then induced, using 4–6 ml/kg of 20% meconium slurry, as before. Before meconium administration, PIP was increased by 3 cm H2O, and respiratory rate set at 25 breaths/minute, and further increases were transiently made as necessary to maintain adequate minute ventilation. After meconium instillation, ventilation was maintained at the premeconium settings for 30 minutes, at which time a blood sample was drawn for arterial blood gas analysis, and indices of lung function were recorded as before. Forty minutes after induction of the meconium injury, PEEP was increased to 6 cm H2O, and animals were randomized (by disclosure of a sealed number) to one of three groups. Animals in the control group received a single pass of dry suction to the tip of the endotracheal tube using a standard 10-FG side-hole suction catheter, simulating the suctioning procedure routinely used in clinical practice to maintain endotracheal tube patency. The volume of meconium slurry recovered was less than 1 ml in all cases. Animals in the DS and PE groups received therapeutic lung lavage, which was limited to a total lavage volume of 30 ml/kg based on the results of the Phase I study. DS lavage fluid was prepared by diluting full-strength bovine surfactant (Survanta; Abbott Australasia, Kurnell, Australia), 1 part in 10 with normal saline, to make a final phospholipid concentration of 2.5 mg/ml. PE fluid was made by mixing Rimar 101 (Mercantile Development Inc., Shelton, CT) 1 part in 5 (wt/vol) with saline, using Survanta as the emulsifier, at a final concentration of 2.5 mg/ml. The phosphatidylcholine in bovine surfactant is known to be an effective emulsifier in the formation of perfluorocarbon–saline emulsions of the oil-in-water type (31). Emulsification was achieved by homogenization for 10 minutes using an Ultra-Turrax homogenizer (model TP18; Janke and Kunkel, Staufen, Germany). By virtue of the density of perfluorocarbon (1.76 g/ml), the emulsion settled over time to a perfluorocarbon-rich lower phase and an upper aqueous phase, as has been previously observed (31). Lavage fluids were prepared immediately before lavage, and warmed to 37°C. In both lavage groups, the two 15-ml/kg aliquots were instilled and recovered in a manner identical to that in Phase I. Changes in physiologic indices (heart rate, mean blood pressure, SaO2) during and after lavage were noted, and PIP was adjusted to maintain adequate chest movement. Ventilatory management for the 5 hours after lavage (or dry suction in the controls) followed a strict protocol, with the following hierarchy of aims: (1) VT to be kept at exactly 8 ml/kg by adjustment of PIP, and, if necessary, inspiratory time; (2) PCO2 to be in the range 40–50 mm Hg by adjustment of respiratory rate; and (3) arterial pH to be > 7.20 by normalizing PCO2 (or administration of NaHCO3 if metabolic acidosis). Changes to ventilatory parameters to achieve these aims were allowed at any time other than within 15 minutes of the next arterial blood gas sampling. PEEP was kept at 6 cm H2O, and FIO2 at 1.0, throughout. Crs, Rrs and results of arterial blood gas analysis were recorded at 20 minutes and then hourly for 5 hours after lavage. Dry suction of the endotracheal tube was performed hourly, with restoration of lung volume after suctioning using three sigh breaths of 1 second duration at 5 cm H2O above the current PIP. At 5 hours, a chest X-ray was taken at maximum inspiration (PIP 35 cm H2O), a further small volume BAL sample was obtained, and a plasma urea sample taken. The chest X-ray was scored for atelectasis, air-trapping, and parenchymal infiltrates by a Pediatric Radiologist (L.C.) unaware of the group randomization. Each element of the score was separately graded 0–4 based on pre-agreed criteria, relating to both the global and regional characteristics of the X-ray (see Table E1 in the online supplement for scoring system). After killing with 80 mg/kg of pentobarbitone (Lethabarb; Virbac Australia, Peakhurst, Australia), histologic examination of the lung was performed as described in the online supplement. Lung sections were assessed by a Pediatric Pathologist (Y.C.) blinded to the randomization, assigning a score of 0–4 for atelectasis, alveolar inflammation, bronchiolar inflammation, and airway plugging with meconium and mucinous debris (Table E2 in the online supplement). An overall score for the lung in each histologic category was calculated.
Phase III Controlled Trial of DS Lavage Using Large and Small Aliquots In the LA group, the 2 x 15 ml/kg lavage was performed exactly as in Phase II. Lavage in the SA group followed the technique described by Lam and Yeung (22), which, at the time of starting these studies, was the only published trial of surfactant lavage therapy in human infants with MAS. The animal remained connected to the ventilator throughout, and the lavage was conducted via an 8-FG suction catheter advanced through a sideport connector so that its tip was positioned in the lower trachea. With the torso positioned alternately right and left side down, aliquots of 3 ml of dilute surfactant were instilled into the lung via the tracheal catheter, and the dead space of the catheter was cleared with air. After a 10-second dwell time, the torso was rolled to the opposite side, and a suction pressure of 150 mm Hg was applied to the tracheal catheter for 10 seconds. Sequential instillation and recovery of 3-ml aliquots continued until total lavage volume reached 30 ml/kg, with all return fluid collected in a mucus trap. Management and investigations after lavage were as for Phase II, except that radiologic scoring was not performed, and extravascular lung water content (32), rather than histologic score, was determined in the postmortem lung specimen.
Analysis of Return Fluid from Therapeutic Lavage
Analysis of BAL Fluid
Statistical Analysis
Phase I Eight piglets of mean weight 4.6 kg (SD 0.53) were studied; all received 4 ml/kg of 20% meconium slurry. Overall 69% of instilled meconium solids and 73% of pigment were recoverable by immediate saline lavage (Figure 1) , highlighting the potential effectiveness of large volume lavage in cleansing the lung. The bulk of meconium recovery occurred with a total lavage volume of 30 ml/kg, and fluid deposition steadily increased, exceeding 10 ml/kg as total lavage volume reached 45 ml/kg. A total lavage volume of 30 ml/kg was thus chosen for Phases II and III of this study, this being a balance between optimization of lung cleansing, and minimization of the amount of aqueous fluid retained in the lung.
Phase II Ten animals were studied in the DS group, and eight in the PE and control groups. Two animals in the PE group and one control animal died of pulmonary hemorrhage before 5 hours after lavage. The impairment of oxygenation and pulmonary mechanics after meconium instillation was similar in all animals (Table E3 in the online supplement). Preliminary experiments revealed that meconium injury was associated with a 40% increase in mean PAP (Figure E1 in the online supplement), with a further increase over the subsequent 5 hours, as has been noted previously in the piglet model of MAS (36). Typical radiographic and histologic findings in nonlavaged control animals are shown in Figures E2 and E3, respectively in the online supplement. The chest X-ray appearances were similar to those in human infants with MAS, with widespread patchy infiltrates, areas of atelectasis, and focal and global hyperinflation. Histologically, the MAS model was dominated by atelectasis, with widespread foci of alveolar collapse. Meconium and mucinous debris were also seen in the airways, along with the beginnings of an inflammatory response. In the DS and PE groups, the 30 ml/kg lavage procedure was associated with transient physiological disturbance, in particular fall in SaO2 (Table 1) , which was most prominent during suctioning. In all cases SaO2 recovered to near-baseline values within a few minutes (maximum 7 minutes). In preliminary experiments on five animals, mean PAP was found to increase by up to 5 mm Hg during DS lavage, but returned to prelavage values shortly after completion. Lavage fluid recovery did not differ significantly between the DS and PE groups, but was somewhat lower than in the Phase I animals, resulting in slightly higher deposition of aqueous fluid into the lung (DS 11 vs. PE 9.3 vs. Phase I 8.0 ml/kg, p = 0.024, ANOVA). Recovery of meconium pigment, but not solids, was significantly higher during DS lavage compared with PE (Table 1). Meconium recovery in both groups was lower than that noted with 30 ml/kg lavage in Phase I, but still more than a third of the instilled meconium was recoverable from the lung when lavage was performed 40 minutes after injury.
A marked improvement in oxygenation (i.e., lower AaDO2) was noted in the DS group after lavage (Figure 2) , with an improvement of lesser degree and duration noted in the PE group. AaDO2 remained lower at 300 minutes postlavage in the DS animals, indicating a sustained effect of lavage in this group (DS: 250 mm Hg; PE: 460 mm Hg; controls: 460 mm Hg; p = 0.0031, ANOVA). Additionally, Crs was better in the DS group during the postlavage period, although not at 300 minutes postlavage. Whereas VT and minute ventilation were similar between the groups throughout the experiment, CO2 was 10–15 mm Hg higher in the PE group than other groups at 2, 3, and 4 hours postlavage. Systolic and mean blood pressures did not differ significantly between the groups at any time point.
The DS lavage group also had less evidence of MAS radiologically compared with the other groups (Figure 3) , with reductions in scores in all three radiologic categories (individual data not shown). Histologically, the group differences were confined to the extent and distribution of atelectasis, which was less prominent in the DS group compared with either the PE group or the controls (Figure 4) . Protein and albumin content in the final BAL sample was lower in the DS group (Table 2) , suggesting a lesser degree of exudative edema than in the control and PE groups.
Phase III Eight animals were studied in each group, with no differences in the amount of meconium instilled, nor in the resultant physiological effects (Table E4 in the online supplement). One animal in the SA group died at 4 hours postlavage of pulmonary hemorrhage. Lavage in the LA group was conducted more rapidly than SA lavage, which took on average 15 minutes to complete (Table 3) . Fall in SaO2 during lavage was more pronounced in the LA group, but was short lived, with recovery to within 5% of baseline values in less than 2 minutes in all cases. With SA lavage, SaO2 remained at approximately 5% below baseline throughout.
Clearance of meconium solids and pigment from the lung was considerably better during LA lavage (Table 3). Proportional recovery of lavage fluid was also higher in the LA group, such that less aqueous fluid, but also less surfactant phospholipid, was deposited into the lung with this form of lavage. Removal of endogenous protein was quite variable, and was not significantly different between the groups, although trended higher in the LA group. Pulmonary mechanics were better in the LA group (Figure 5) , both throughout the entire postlavage ventilation period, and also at 5 hours postlavage (at 5 hours, Crs: LA 0.69 vs. SA 0.50 ml/cm H2O/kg, p = 0.019; Rrs: LA 95 vs. SA 121 cm H2O/L/second, p = 0.013). AaDO2 trended lower throughout the postlavage period in LA animals (p = 0.067, repeated measures ANOVA), and was significantly lower at 5 hours postlavage (p = 0.046). Systolic and mean blood pressures did not differ significantly between the groups at any time point.
There was evidence of more marked proteinaceous edema in the SA group, with higher protein and albumin content in the final BAL sample (Table 4) , and a higher extravascular lung water content (Table 5) .
In these studies we found that a total lavage volume of 30 ml/kg removed considerable amounts of meconium from the lung, while keeping fluid deposition below 10 ml/kg. Using this total lavage volume, we then found that dilute surfactant was a very effective lavage fluid in experimental MAS, most notably when administered in two aliquots of 15 ml/kg. The piglet model of MAS used in this experimental series demonstrated several of the cardinal features of human MAS (1, 2), including pronounced and prolonged disturbance of gas exchange, impaired pulmonary mechanics, proteinaceous and hemorrhagic edema, and radiographic infiltrates. Control animals showed severe lung injury macroscopically and histologically, with even distribution throughout the lung lobes. Clearly, the induction of MAS by instillation of meconium slurry into the air-filled lung of a 2-week-old piglet excludes several key elements of the pathogenesis of the disease in humans, in particular fetal hypoxia, and inhalation of meconium into a fluid-filled lung (1, 2). Human infants with MAS frequently demonstrate labile pulmonary hypertension, with significant muscularization of pulmonary arterioles in advanced cases (37). The piglet model of MAS does exhibit a degree of pulmonary hypertension (36), but without the lability and reactivity so prominent in many neonates with MAS. The findings of this and any other study involving an animal model of MAS must be interpreted in this light. Our studies have identified a protocol for lung lavage that appears to effectively clear meconium from the piglet lung, with resultant improvement in lung function. Whether this lavage sequence will be well tolerated and effective in human infants cannot be determined from experiments in an animal model, and will require further investigation. The total lavage volume of 30 ml/kg that we found optimal in Phase I is greater than that used by most investigators for lavage of human infants with MAS, with the exception of Mosca and coworkers (30 ml/kg saline in two aliquots) (21) and Wiswell and colleagues (48 ml/kg in six aliquots) (27). There is strong evidence, both from our investigations and those of others, that a relatively large volume of lavage fluid is required to cleanse the lung effectively in MAS, particularly where meconium has dispersed into the distal airspaces. With immediate lavage in Phase I, there was a considerable increment in meconium recovery between 15 and 30 ml/kg lavage, suggesting that a total lavage volume lower than the functional residual capacity (around 30 ml/kg in MAS [38]) may not cleanse the lung adequately. Similarly, in the rabbit model of MAS, Cochrane and coworkers (11) also found that high lavage volumes were required to recover meconium, with pigment recovery reaching 36% once 40 ml/kg of lavage fluid had been instilled. On the other hand, lavage volumes beyond 30 ml/kg did not recover significant additional quantities of meconium in our study; nor did volumes greater than 40 ml/kg in the rabbit model of MAS (11). Meconium recovery during lavage was also affected by aliquot volume, with recovery of both meconium solids and pigment being higher using aliquots of 15 ml/kg. These data suggest that there was greater permeation of lavage fluid down the tracheobronchial tree with LA lavage, such that removal of meconium from the distal airspaces was enhanced. Previous studies of the distribution of fluid introduced into the airways clearly indicate that larger instilled volumes disperse more peripherally in the tracheobronchial tree. During lavage of a segment of the right middle lobe in adults, Kelly and colleagues (39) found that distal dispersion only occurred after instillation of at least 120 ml of lavage fluid. Conversely, small fluid boluses (< 5 ml) introduced into the trachea in adult dogs show poor distal dispersion, with fluid remaining predominantly in the main bronchi (40). Larger aliquot volume may also achieve better meconium clearance because the instilled lavage fluid is more evenly distributed throughout the lung. Van der Bleek and coworkers (41) found in surfactant-depleted rabbits that rapid instillation of a 16-ml/kg bolus of bovine surfactant (100 mg/kg) resulted in more homogeneous distribution within the lung lobes than when the surfactant was given in a volume of 2.4 ml/kg. The passive deflation of the lung after disconnection from the ventilator before lavage in the LA group may have also enhanced fluid distribution. Homogeneity of distribution of instilled fluid is known to be higher in a low volume fluid-filled lung than a well expanded lung (42). The chest compression and vibratory massage during the suctioning phase of the lavage procedure may also have aided meconium recovery in the LA group. As with meconium recovery, fluid return during lung lavage is dependent both on total lavage volume and aliquot volume, with larger volumes yielding greater proportional recovery. Proportional fluid recovery in Phase I increased from 68% for a 15 ml/kg lavage to 82% for a total lavage volume of 60 ml/kg. More modest fluid returns have been noted in studies of lavage in human infants, ranging from 34% (22) to 76% (21). Using a total lavage volume of 48 ml/kg, Wiswell and coworkers (27) reported a mean fluid recovery of only 51%, meaning that more than 20 ml/kg of aqueous fluid remained in the lung after lavage. In Phase III of the present study, proportional fluid recovery was also found to depend on aliquot volume, with an additional 11% fluid return using aliquots of 15 ml/kg compared with 3 ml. This finding concurs with the limited data available in human infants, where proportional fluid recovery has been reported to be 34% with 2-ml aliquots (22), 51% with aliquots of 8 ml/kg (27), and 76% with an aliquot volume of 15 ml/kg (21) Deposition of aqueous fluid in the airspaces is an inevitable consequence of therapeutic lung lavage. The potential deleterious effects related to this transient alveolar flooding have recently been expounded (43), and are of concern in a seriously ill ventilated infant with MAS. For this reason, we chose a total lavage volume in Phase II that limited aqueous fluid deposition to around 10 ml/kg. It must be recognized that the pulmonary epithelium is highly permeable to water (44), and the newborn lung in particular is uniquely equipped to clear fluid, by virtue of the very high activity of sodium transport pathways (45). In adults with alveolar proteinosis, the lung has been found capable of very rapid alveolar fluid clearance after large volume therapeutic lavage (46). No equivalent data are available in infants, but the physiologic data in this and other animal studies, and the reported experience in human infants with MAS, do suggest that the aqueous fluid deposited in the lung during lavage clears relatively quickly. The net deposition of surfactant phospholipid that was noted in all groups during lavage may have aided in the clearance of fluid from the alveolar space (47). Both DS and PE lavage in Phase II led to improvements in oxygenation in the post-lavage period that were not noted in controls. Only in the DS group, however, was there a sustained effect on oxygenation, associated with evidence of reduced lung injury at 5 hours post-lavage. This occurred despite application of a rigorous protocol for ventilatory management in all groups, ensuring that intergroup differences could not be explained by discrepancies in ventilator settings. The improved lung function in the DS group highlights the potential efficacy of DS lavage in MAS, and is complementary to the findings of previous investigations in animal models of MAS (9–11, 13, 17–19). The PE fluid used in Phase II has the same proportions of perfluorocarbon to aqueous fluid as have been used in a blood substitute emulsion (Fluosol DA, Green Cross, Osaka, Japan) (48). It is important to consider why the PE fluid, with a higher oxygen-carrying capacity and lower surface tension than DS fluid, did not confer any benefit. Recovery of meconium pigment, but not solids, was somewhat lower with the PE fluid, possibly limiting the potential benefit of lavage in this group. It is also possible that, in its role as an emulsifier, the surfactant phospholipid in the PE lavage fluid was "trapped" at the interface between perfluorocarbon and saline, with limited capacity to act in overcoming endogenous surfactant inhibition or alveolar fluid clearance. Finally, the deterioration in both oxygenation and CO2 clearance several hours after PE lavage may be related to regional hypoventilation secondary to airway plugging with a gelatinous secondary emulsion rich in surfactant and perfluorocarbon, which has been noted by other investigators to accumulate in the lung after perfluorocarbon instillation (49). In Phase III, we found that post-lavage lung function was significantly better using aliquots of 15 ml/kg compared with 3 ml, with sustained improvement in AaDO2, Crs, and Rrs. All evidence points to this being related to the increased clearance of meconium from the lung in this group. It is possible that the extra deposition of aqueous fluid during lavage in the SA group may have contributed to lung dysfunction in the early postlavage period, but not at 5 hours beyond lavage. Endogenous protein recovery trended higher with LA lavage, but again this is not the likely explanation of the differences in pulmonary mechanics and oxygenation seen after 5 hours, and cannot be implicated as the cause of reduced lung injury. This study therefore emphasizes that the primary objective of lung lavage in MAS is to remove meconium from the lung, and that, by doing so, significant improvement in lung function may follow. Secondary benefits of surfactant lavage in MAS may include removal of proteinaceous exudate and mucinous debris, and relatively homogeneous distribution of a dose of exogenous surfactant.
CONCLUSIONS
The authors thank Abbott Australasia for provision of surfactant; Chiron Diagnostic for loan of the arterial blood gas analyser; and Mr. Magdy Sourial, Ms. Julie-Anne Frederiksen, and Ms. Wendy Russell for technical assistance.
This study was supported by Murdoch Childrens Research Institute (Project Grant No. 98007 and Career Development Grant No. 97009) and by Abbott Australasia (grant-in-aid and provision of surfactant). This article has an online data supplement which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form January 28, 2003; accepted in final form April 20, 2003
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
Crs values lower in DS group through postlavage period (p = 0.00021), but not significantly different at 300 minutes (p = 0.087). Rrs values were not significantly different between groups (p = 0.34 and 0.46, repectively).
