INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary surfactant extracts, and surfactant substitutes, are remarkably successful in treating respiratory distress syndrome of premature infants (1), a disease characterized by a deficiency of pulmonary surfactant (2). However, effects of surfactant treatment are less dramatic when surfactant is used to treat lung diseases associated with acute lung injury (3). Insufficient distribution, insufficient dosage, advanced lung damage, and inactivation have been suggested to explain poor response to treatment.

Tierney and Johnson (4) were among the first to suggest that inactivation of pulmonary surfactant may play a role in human disease. Causes of inactivation are multiple. Serum (5), plasma proteins (6), hemoglobin (7), proteases (8), phospholipases (9), hyperoxia (10), antibodies (11), pH and cations (12, 13), lipids (14), meconium (15), and bilirubin (16) all interfere with surfactant function (17).

A number of studies have shown that infants with meconium aspiration pneumonia respond modestly to surfactant treatment (18, 19). Meconium is a potent inactivator of surfactant function in vitro, and yet we have found that mixtures of Survanta with polymers maintain good surface activity in the presence of meconium and other inactivating substances (20). We therefore carried out a series of experiments to test if polyethylene glycol (PEG) or dextran added to surfactant improves serial measures of pulmonary function in a rat model of meconium aspiration pneumonia.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Adult Sprague Dawley rats of either sex and weighing 250 to 400 g were anesthetized by injecting pentobarbital, 60 mg/kg, intraperitoneally. After tracheostomy, animals were supported on a Harvard volume-controlled ventilator (Harvard small rodent ventilator, model 683; Harvard Apparatus, South Natick, MA) with initial settings of frequency, 65 breaths/min; tidal volume, 9 ml/kg body weight; positive end-expiratory pressure, 4 cm H₂O; FI_O2, 1; and flow, 0.5 L/min. Pancuronium (1 mg/kg) was used for paralysis. The right carotid artery was catheterized for blood pressure measurements, sampling of blood gases, and drug and fluid infusion. Ten units of heparin per ml of saline were used to flush the catheter. Blood and tracheal pressures were measured using Viggo-Spectromed transducers (Gould Model P23XL) attached to a recorder (Gould Windograf; Gould, Valley View, OH). Manometers were used to calibrate both pressure transducers. Throughout the experiment, ventilation was adjusted to maintain Pa_CO2 between 40 and 55 mm Hg, first by changing the frequency (range, 50 to 70/min) and then by adjusting the tidal volume according to the adequacy of chest excursions.

First-passed meconium was taken from urine-free diapers of normal term infants in our well-baby nursery. The meconium was placed in a sterile jar and frozen for < 7 d and then lyophilized. Pooling meconium from three to six infants minimized possible variability. Dry meconium was mixed in sterile water by Vortex at a concentration of 30 mg/ml, then aspirated into a tuberculin syringe through a 25-gauge needle (21). Five ml/kg body weight was instilled via the tracheostomy in less than 10 s. Half the dose was given with the rat lying on one side and half with the rat lying on the other side. Five minutes after giving the meconium, furosemide, 10 mg/kg, was injected via the carotid artery to reduce fluid retention in the lungs (21).

Blood gas determinations, systolic blood pressure, tidal volume, and peak inspiratory pressures were recorded approximately 15 min after the meconium injury. Repeat doses of meconium were administered (1 ml/kg) if the Pa_O2 was above 115 mm Hg.

Sixty minutes after the last dose of meconium, treatment was given with the rats positioned on the right then the on left side during each half dose. Survanta was used for treatment because of its availability and because of its prior clinical use for meconium aspiration pneumonia (18). It was obtained courtesy of Ross Laboratories (Columbus, OH) or from excess after treatment of newborns. Prior to use in these experiments, the surface tension of Survanta was tested in a modified pulsating bubble surfactometer (Electronetics, Buffalo, NY) at a concentration of 1.25 mg/ml (20). Minimal surface tension in the pulsating bubble surfactometer was assessed at 37° C on the tenth cycle and in all cases was < 10 mN/m.

Polyethylene glycol (PEG, with a molecular weight of 10 kD, Lot 115H26011) and dextran (MW of 9.5 kD, Lot 77H1253) were obtained from Sigma Chemical (St. Louis, MO). PEG or dextran, 5% (wt/vol), was added to Survanta, then mixed by Vortex for as long as 30 s. We used standard treatment volumes of 4 ml/kg of body weight.

The experimental groups are shown in Table 1. Groups 1 to 3 had no meconium injury and received either no instillation, PEG (5%; 10 kD) in normal saline, or normal saline, respectively. Groups 4 to 6 had meconium injury and were either not treated, treated with normal saline, or treated with PEG (5%; 10 kD) in normal saline. Groups 7 to 11 were injured with meconium, then received various doses of Survanta, Survanta with PEG, or Survanta with dextran.

Blood gas determinations, blood pressure, tidal volume, ventilator rate, and peak inspiratory pressure were recorded 30 min after treatments, then hourly. Specific compliance (Crs) of the respiratory system was measured using the formula: Crs = [(VT)/[(PIP  PEEP)]  WL where VT is the tidal volume (ml), PIP is the peak inspiratory pressure (cm H₂O), PEEP is the positive end-expiratory pressure, and, WL is the predicted weight of the lungs (0.8% of body weight).

Three hours after treatment, pentobarbital was injected and the lungs degassed by clamping the trachea. The abdomen and diaphragm were opened and the vena cava cut. Postmortem quasi-static deflation pressure-volume measurements were carried out by inflating the lungs to a pressure of 35 cm H₂O. After a period of stabilization (about 45 s or when < 0.1 ml of air entered in 10 s), volume was measured to define total lung capacity. Pressure was reduced in steps of 5 cm H₂O with at least a 10-s stabilization at each step and the corresponding volumes measured. Volume was corrected for compression in the dead space of the apparatus.

The lungs were then removed, blotted, and weighed. A whole midsagittal slice of right lung was taken for histology and the remaining lungs were weighed wet then again dry after 24 h at 50° C. To compare the increase in lung weight relative to the volume of meconium and treatments, we determined a ratio: measured lung weight minus predicted lung weight/volume instilled.

The right lung segment was placed in 4% formalin, embedded in paraffin, and stained with periodic acid-Schiff reagent (PAS). The sections were coded so that the pathologist (BR) graded the specimens without knowledge of the experimental groupings. The sections were examined by light microscopy and assessed for the presence of (PAS-positive) meconium in the airways, intra-alveolar edema, hemorrhage, hyaline membranes, and leukocytes. Each characteristic was scored 0 to 3 (0 = absent; 1= mild; 2 = moderate; 3 = prominent).

Data are expressed as mean ± SEM. Serial measurements were analyzed by one- or two-way analysis of variance (ANOVA) modified for repeated measures using SigmaStat software (SPSS Science, Chicago, IL). When intergroup differences occurred, these differences were further analyzed comparing the two groups of interest; p < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

General State of Animals

Despite the heterogeneity of meconium, this form of lung injury resulted in animals having a significant (and stable) degree of alveolar arterial oxygen difference (Figure 1), whereas blood pressure and acid-base status, in general, were maintained for as long as 4 h (Table 1). Sixty-six out of seventy animals survived for the duration of the protocol. No pneumothoraces occurred.

View larger version (16K): [in this window] [in a new window]   Figure 1.   Serial measures of PaO2 are shown for the experimental groups. (Top panel ) Results for the experimental groups listed in Table 1 are shown. None of these groups received meconium injury. The diamonds represent means ± SEM for the group that was ventilated with no tracheal instillation. Squares indicate the groups instilled with saline, and triangles indicate the group instilled with PEG (Groups 1, 2, and 3 in Table 1). The horizontal axis indicates the preinstallation and postinstillation times for the two groups that received intratracheal instillation. (Middle panel ) Average values are shown for groups that were injured with meconium but received no surfactant treatment. Diamonds indicate meconium injury alone; squares indicate meconium injury followed by saline sham treatment; triangles indicate meconium injury followed by PEG sham treatment. (Groups 4, 5, and 6 in Table 1). (Bottom panel ) Results for the five groups that received surfactant (with or without PEG or dextran) after meconium injury (7 to 11 in Table 1) are shown. Circles indicate average values for the group treated with Survanta 35 mg/kg with PEG; the thick bar is Survanta 50 mg/kg with PEG; the diamonds are Survanta 50 mg/kg with dextran; the squares are Survanta 50 mg/kg; the triangles are Survanta 100 mg/kg. Note that the group treated with Survanta 50 mg/kg with PEG is significantly different from both injured control animals and animals treated with either 50 or 100 mg/kg of Survanta alone (two-way ANOVA with repeated measures, p < 0.05).

Experimental Groups 1 to 3 Normal

These animals were not injured with meconium. There were no significant differences between the three groups with regard to serial measures of oxygenation, peak inspiratory pressures, Pa_CO2, metabolic acidosis, blood pressure, or lung weights (Figures 1 and 2 and Tables 1 and 2). Pressure volume curves did not differ except that the group that received intratracheal saline had an average total lung capacity 13% higher than the group that received no instillation (p < 0.05) (Figure 3). Lung weights are shown in Table 2. Lung weights for normal animals averaged 0.8% of body weight, which is not different from the average value we obtained for a group of unventilated rats (data not shown).

View larger version (19K): [in this window] [in a new window]   Figure 2.   Serial measures of peak inspiratory airway pressures are shown for the experimental groups. The symbols for the experimental groups in the top, middle, and bottom panels are the same as Figure 1. The group treated with Survanta 50 mg/kg with PEG (thick line) is not significantly different from the normal groups in the top panel. Also this group (Survanta 50 mg/kg with PEG) is significantly different from both injured control animals and animals treated with either 50 or 100 mg/kg of Survanta alone (two-way ANOVA with repeated measures, p < 0.002).

View larger version (18K): [in this window] [in a new window]   Figure 3.   Average deflation pressure-volume relationships for the 11 experimental groups are shown. Symbols for experimental groups are the same as those in Figure 1. The group treated with Survanta 50 mg/ kg with PEG (thick line) is not significantly different from the noninjured control group, and is significantly different from the group treated with Survanta 50 mg/kg (two-way ANOVA with repeated measures, p < 0.03).

Histology of lungs from three rats that received saline instillation and from three rats that received ventilator support with no tracheal instillation were examined. All these lungs were rated as normal (Figure 4).

View larger version (149K): [in this window] [in a new window]   View larger version (153K): [in this window] [in a new window]   View larger version (153K): [in this window] [in a new window]   View larger version (112K): [in this window] [in a new window]   Figure 4.   Histology. (A) Low power view showing widespread inflammatory lung injury with prominent hyaline membranes in a lung from an animal treated with Survanta 50 mg/kg. Aspirated meconium can be seen as dark-stained material in terminal air spaces (e.g., arrow). PAS × 100. (B) Detail from an animal that was injured with meconium and received no surfactant. Inflammatory lung injury is evident with prominent hyaline membranes. Aspirated meconium is indicated by arrows. PAS × 100. (C ) Inflammatory injury in an animal that has been treated with Survanta 50 mg/kg with PEG. Meconium in a dilated bronchiole is indicated by the arrow. No hyaline membranes are seen. PAS × 100. (D) Normal lung parenchyma is seen in the lungs of an animal that was not injured with meconium. PAS × 100.

Experimental Groups 4 to 6: Injury, No Surfactant Treatment

Groups 4 to 6 all had meconium injury and either received no subsequent instillation, or were instilled with normal saline, or with PEG (5%; 10 kD) in normal saline. These groups were similar to each other with regard to serial measures for oxygenation, peak inspiratory pressures, Pa_CO2, metabolic acidosis, and lung weights (Tables 1 and 2). Values for Pa_O2 remained about 20% of those for the first three (normal) groups (Figure 1). Peak inspiratory pressures were threefold higher than those for the normal groups (Figure 2). Pressure volume curves were shifted downwards from those of the normal groups, and total lung capacity averaged 30% less (Figure 3). Lung weights for these animals were all between 1 and 2% of body weight. In general, the sickest animals in these three groups (highest base deficit, PIP, and Pa_CO2; lowest Pa_O2 and blood pressure) were instilled with the highest volumes of meconium, saline, or PEG. Two animals in the group that received saline and two animals in the group that received PEG died with progressive mixed acidosis and hypotension about 2 h after these fluids were instilled.

Experimental Groups 7 to 11: Injury, Surfactant, or Surfactant/Polymer Treatment

Treatment with PEG/Survanta (50 mg/kg) gave better results than treatment with 50 or 100 mg/kg of Survanta alone in terms of oxygenation, PIP, and pressure-volume curves (p < 0.05). The PEG/Survanta 50 mg/kg group had mean values of blood pressure, pH, Pa_CO2, and base deficits that were not different from those of animals in the noninjured control group. For the five groups that received Survanta (7 to 11), two-way ANOVA for repeated measures indicated that PEG/Survanta 50 mg/kg was the only group in which oxygenation and PIP were significantly different (p < 0.05) from those of animals that received no treatment after meconium injury (Figures 1 and 2).

Although Survanta 50 mg/kg with dextran did not improve oxygenation or reduce PIP significantly when compared with injured animals not receiving surfactant, this group was not significantly different from the normal groups in terms of pressure-volume relationships. The group that received Survanta 35 mg/kg with PEG also had no significant increase in oxygenation or decrease in PIP, but also pressure-volume curves were not different from those of the normal control groups (Figures 2 and 3).

Because we occasionally altered the volume-regulated ventilator to maintain the desired range of Pa_CO2 in all animals, we also calculated dynamic compliance (V/P) for each point. Differences between compliance measures were similar to the PIP differences. For example, average compliance of the Survanta 50 mg/kg group versus PEG/Survanta 50 mg/kg was 0.37 ml/kg/cm H₂O versus 0.69 ml/kg/cm H₂O (p < 0.001).

Two-way ANOVA for repeated measures found that average pressure-volume curves for PEG/Survanta 50 mg/kg were not significantly different from those of the noninjured control group (Group 1) (Figure 3). Average deflation curves showed a significant difference between the PEG/Survanta 50 mg/kg group and the Survanta 50 mg/kg group (p < 0.05), but the differences between PEG/Survanta 50 mg/kg and Survanta 100 mg/kg were not statistically significant.

Lung Weights

Lung weights correlated with the volumes instilled into the animals (p < 0.001) (Table 2). We analyzed whether some groups had lung water in excess of the volumes instilled by computing the actual lung weight minus predicted lung weight divided by the total instilled volume. These results are shown in the last column (Ratio) in Table 2. In no case was the ratio > 1, although three of the four polymer/surfactant groups had the highest ratios.

Histology

Histologic examination revealed the presence of meconium in all animals in which it was instilled (Figure 4). Granulocytic infiltrates were present in all animals that received meconium. Hyaline membranes were present in five of six animals receiving no treatment and in three of six animals treated with Survanta alone, but were not seen in lungs of animals treated with PEG/Survanta (p < 0.01, chi-square). Alveolar edema and hemorrhage were inconsistent findings in all groups but tended to be less prominent in animals treated with PEG/Survanta.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PEG/surfactant mixtures have remarkable effects on lung function in animals with meconium lung injury. Treatment with PEG/Survanta, 50 mg/kg, resulted in improvements in lung compliance and oxygenation that surpassed those found with Survanta alone at either 50 or 100 mg/kg. The results are consistent with in vitro findings that addition of PEG to surfactant prevents/reverses inactivation of Survanta by meconium and other substances (20). Surfactant proteins, particularly SP-A and SP-B, also reduce inactivation of surfactant in a number of in vitro and in vivo models (22, 23). Therefore PEG mimics this function of surfactant proteins. It is not known whether PEG would have similar effects on "natural" surfactants that contain the four known surfactant proteins.

Is the beneficial effect limited to PEG or is it a general phenomenon for nonionic polymers? In earlier work, we found that dextrans, polyvinylpyrrolidones, and PEG all improved function of Survanta in vitro, especially when various inactivating substances were added to the surfactant mixtures (20). Kobayashi and coworkers (24) have found that dextran (71 kD) added to an extract of porcine surfactant improved surface adsorption rates and minimum surface tensions when albumin was added as an inactivating substance. They also found that dextran improved lung function when added to surfactant/albumin mixtures given to premature rabbits. We did not find that dextran/Survanta mixtures in comparison with Survanta alone improved gas exchange or reduced positive inspiratory pressures, although there is a suggestion from pressure-volume measurements of an improved response. Meconium aspiration may represent a more potent form of inactivation of surfactant than occurs in the albumin/rabbit model.

Mechanisms that would explain polymer effects on surfactant in the presence or absence of inactivating substances are uncertain. PEG binds water molecules and excludes other large molecules from the bound water (25). Polymers in concentrations of 5% or greater aggregate lipids, a phenomenon in part caused by exclusion of polymer between bilayers setting up osmotic gradients that dehydrate lipids and cause phase separations (25). We have found that increasing aggregation of Survanta in the presence of increasing concentrations of PEG (0 to 10%) is associated with improved surface activity (28). While these studies are indicative, ways in which PEG or dextrans prevent or reverse different types of surfactant inactivation are doubtless multiple.

Because PEG is a polymer of ethylene glycol, safety of PEG for human use is of obvious concern. Although low molecular weight PEGs (< 400) may have some toxicity, the Federal Drug Administration has approved PEGs of higher molecular weight for internal human consumption. PEG is used for compounding many drugs and cosmetic products and has been investigated (bound with free hemoglobin) as a blood substitute (29).

There are several limitations to these experiments. We have used only one form of lung injury in a single species, we have utilized just one surfactant, we have studied the animals for a relatively short period of time, and we have used only a few measures of lung function. Although we infer that the beneficial effects of PEG in this model of lung injury are linked to its ability to prevent or reverse surfactant inactivation, we can only surmise that this in vitro effect is responsible for the in vivo results.

In conclusion, PEG added to Survanta remarkably improves lung function in animals with acute lung injury caused by meconium. A number of clinical implications are obvious if these findings are confirmed and amplified. Foremost is the possibility that existing formulations of therapeutic surfactants can be improved with simple modifications. These modifications may provide enhanced resistance to inactivation. If inactivation of surfactant plays a major role in various pulmonary diseases involving acute lung injury (e.g., adult respiratory distress syndrome), then polymers and/or surfactants containing polymers may improve therapy.