Abstract
The efficacy of exogenous surfactant administration is influenced by numerous factors, which has resulted in variable outcomes of clinical trials evaluating this treatment for the acute respiratory distress syndrome (ARDS). We investigated several of these factors in an animal model of acid aspiration including different surfactant preparations, and different delivery methods. In addition, high-frequency oscillation (HFO), a mode of mechanical ventilation known to recruit severely damaged lungs, was utilized. Lung injury was induced in adult rabbits via intratracheal instillation of 0.2 N HCl followed by conventional mechanical ventilation (CMV) until PaO2 /Fi O2 values ranged from 220 to 270 mm Hg. Subsequently, animals were given one of three surfactants administered via three different methods and physiological responses were assessed over a 1-h period. Regardless of the surfactant treatment strategy utilized, oxygenation responses were not sustained. In contrast, HFO resulted in a superior response compared with all surfactant treatment strategies involving CMV. The deterioration in physiological parameters after surfactant treatment was likely due to overwhelming protein inhibition of the surfactant. In conclusion, various surfactant treatment strategies were not effective in this model of lung injury, although the lungs of these animals were recruitable with HFO, as reflected by the acute and sustained oxygenation improvements.
The acute respiratory distress syndrome (ARDS) is characterized by hypoxemia, decreased lung compliance, and pulmonary edema, and is associated with a mortality ranging from 40 to 60% (1). ARDS results from a number of direct or indirect pulmonary insults including aspiration, trauma, near-drowning, and sepsis (2). Gastric acid aspiration is a common etiology of ARDS and manifests initially by a chemical “burn” to the pulmonary epithelium, with a subsequent influx of edema fluid into the alveolar space. This results in surfactant inactivation, and significant atelectasis (3).
Pulmonary surfactant is composed of 90% phospholipid and 10% specific surfactant-associated proteins designated as SP-A, -B, -C, and -D (4). Surfactant functions to reduce surface tension at the air–liquid interface, thus maintaining alveolar stability. Because alterations of the endogenous surfactant system contribute to the lung dysfunction associated with ARDS (5, 6), administration of exogenous surfactant is currently being evaluated in these patients. Clinical trials conducted to date, however, have reported inconsistent results (7, 8). This is due, in part, to the complex pathophysiology associated with ARDS, and the observation that multiple factors can affect the efficacy of this treatment modality. Such factors include the specific surfactant preparation utilized (9), the dose and delivery method of the exogenous surfactant (10), as well as the mode of mechanical ventilation utilized during and subsequent to surfactant administration (11, 12). In addition, one report has shown that the nature and severity of the underlying lung injury may also influence the host's response to exogenous surfactant administration (13). Given the importance of these factors influencing the response to surfactant treatment, it is essential to evaluate and compare different treatment strategies in animal models of lung injury before initiating such approaches in the clinical setting. On the basis of current experimental evidence, it is likely that the specific etiologic factor predisposing a patient to ARDS, such as gastric aspiration, will require a particular surfactant treatment strategy tailored to that patient's needs.
The purpose of the current study was to evaluate various exogenous surfactant treatment regimens involving different surfactant preparations (synthetic versus natural), and different surfactant delivery methods (lung lavage versus instillation), in a rabbit model of gastric acid aspiration. In addition, a relatively novel mode of mechanical ventilation (high-frequency oscillation) was also evaluated in this model to determine whether these lungs were recruitable.
Three exogenous surfactant preparations (natural, bovine lipid extract surfactant [bLES], and recombinant SP-C [rSP-C]) with similar phospholipid concentrations, were used in this study. Natural ovine surfactant containing all the surfactant lipids and the surfactant-associated proteins was obtained via whole lung lavages of adult sheep. The recovered alveolar lavage was centrifuged initially at 150 × g to remove cellular debris, followed by centrifugation at 40,000 × g for 15 min at 4° C to obtain a pellet corresponding to the large surfactant aggregate (LA) fraction. This pellet was resuspended in 0.15 M NaCl to a final concentration of 25 mg/ml. Bovine lipid extract surfactant (bLES) (bLES Biochemicals, London, ON, Canada) is a modified natural surfactant preparation extracted with organic solvents from bovine alveolar lavage. bLES is composed of surfactant phospholipids as well as the hydrophobic proteins SP-B and SP-C. The synthetic, recombinant SP-C-based surfactant (Venticute; Byk Gulden, Konstanz, Germany) is composed of 2% human recombinant SP-C protein, 93% phospholipids (dipalmitoylphosphatidylcholine and dipalmitoyloleophosphatidylglycerol in a 70:30 [wt/wt] ratio), and 5% palmitic acid. Venticute was reconstituted from a lyophilized powder in 0.15 M NaCl to a final concentration of 25 mg/ml.
Adult female New Zealand White rabbits weighing 2.0–2.5 kg were sedated with an intramuscular injection of ketamine hydrochloride (20 mg/kg) and acepromazine (0.5 mg/kg). Animals were placed in the supine position and 1% lidocaine was injected subcutaneously over the region of the trachea. The trachea was exposed and a tracheotomy was performed by inserting an endotracheal tube (3.5–4.0 mm) with a sideport adaptor. The carotid artery was cannulated by securing a 5.0 F feeding catheter to allow for blood gas measurements, saline and anesthetic infusion, and the continuous monitoring of arterial blood pressure via a pressure transducer and monitor (Hewlett-Packard, Palo Alto, CA). Animals were subsequently anesthetized via a carotid artery infusion with sodium thiopental (5–10 mg/kg), followed by pancuronium bromide (0.05 mg/kg) to inhibit spontaneous breathing. These anesthetic agents were administered intermittently throughout the duration of the experiment to maintain an adequate level of anesthesia. Animals were ventilated in the prone position, using a pressure-limited infant ventilator (model IV-100B; Sechrist, Anaheim, CA) with the following parameters: a fraction of inspired oxygen (Fi O2 ) of 1.0, a respiratory rate of 30 breaths/min, a tidal volume (Vt) of 8–10 ml/kg, a positive end-expiratory pressure (PEEP) of 5 cm H2O, and an inspiratory:expiratory ratio of 1:1. An initial Vt recording was performed by connecting a pneumotachometer (Hans Rudolph, Kansas City, MO) between the endotracheal tube and the ventilatory circuit. The Vt was adjusted by altering the peak inspiratory pressure (PIP) to maintain a volume of 8–10 ml/kg. An arterial blood gas measurement was performed with a blood gas analyzer (ABL-500; Radiometer, Copenhagen, Denmark) after a 10-min stabilization period in order to establish baseline values.
Immediately after baseline parameters were obtained, 0.2 N hydrochloric acid (HCl) was instilled through a sideport adaptor of the endotracheal tube in a volume of 4 ml/kg during the inspiratory phase of respiration. The acid was administered in two aliquots, with half the volume delivered with the animal on its right side and the remainder with the animal placed on its left side. After the acid instillation procedure, an air bolus of 5–10 ml/kg was injected through the sideport to allow for peripheral distribution of the acid, and PIP was increased by 4–6 cm H2O to maintain adequate chest movement based on visual assessments. Vt was subsequently recorded and adjusted to 8–10 ml/kg every 15 min and blood gases were monitored at 5, 10, and 15 min and every 15 min thereafter until the PaO2 /Fi O2 value was between 220 and 270 mm Hg. This range of PaO2 /Fi O2 values corresponded to 50% of the baseline values of these animals and occurred at approximately the 45-min time point after the acid was instilled. Previous studies have shown that this oxygenation criterion resulted in adequate animal viability throughout the entire experimental period for this model (13). Animals meeting this inclusion criterion were then entered into one of the following four treatment groups, or ventilated for 60 min to serve as untreated, acid control animals.
Surfactant instillation. Animals fulfilling appropriate inclusion criteria after HCl instillation were given one of the three exogenous surfactant preparations (bLES, rSP-C, or natural ovine surfactant) as a liquid bolus into their lungs. A dose of 100 mg of phospholipid per kilogram body weight in a concentration of 25 mg/ml was used. The surfactant was administered intratracheally in a volume of 4 ml/kg, using the same method described for acid instillation, including the increase in PIP values of 4–6 cm H2O immediately after the administration of surfactant in order to maintain adequate chest movement. Blood gases were monitored as specified above and the PIP adjusted every 15 min to maintain a Vt of 8–10 ml/kg over the 60 min of ventilation after treatment.
Surfactant lavage. Separate groups of animals were utilized for this treatment approach, which involved surfactant administration by a whole lung lavage procedure. Each of the three surfactant preparations was diluted with 0.15 M NaCl at 37° C to obtain a final concentration of 4.5 mg/ml. A total dose of 100 mg of phospholipid per kilogram body weight was utilized, which resulted in a lavage volume of 22 ml/kg body weight. Animals meeting the inclusion criterion after HCl instillation were disconnected from the ventilator and one of the dilute surfactant suspensions (or 22 ml/kg of 0.15 M NaCl) was infused into the lungs via the endotracheal tube and withdrawn slowly. Animals were then reconnected to the ventilator and the PIP was immediately increased by 4–6 cm H2O to maintain adequate chest movement. This procedure took approximately 40 s and was well tolerated by the animals with only a slight and transient decrease in arterial pressure observed, with subsequent stabilization. The recovered volume of lavage material from this procedure, designated as Lavage 1, was recorded and stored for further analysis. Blood gases were monitored and Vt adjusted over the 60-min period as described previously.
Saline lavage followed by surfactant instillation. Animals fulfilling the inclusion criteria after HCl instillation were disconnected from the ventilator and 22 ml/kg body weight of 0.15 M NaCl at 37° C was infused through the endotracheal tube into the lungs and subsequently withdrawn as described above. The recovered volume of Lavage 1 was recorded and stored for further analysis. Immediately after this saline lavage procedure, one of the three exogenous surfactant preparations (100 mg of phospholipid per kilogram body weight in a 4-ml/kg liquid bolus) was instilled through the endotracheal sideport adaptor as described above. Animals were then reconnected to the ventilator with an immediate increase in PIP of 4–6 cm H2O. The total time required for this procedure was approximately 50 s. Similar to the previous surfactant lavage groups, this procedure was well tolerated by all animals, with no significant hemodynamic changes observed apart from a transient decrease in arterial pressure. Animals were subsequently monitored via blood gas analyses and Vt measurements as described above.
High frequency oscillation. Animals meeting inclusion criteria after HCl instillation were immediately connected to a high-frequency oscillator (model 3100A; Sensormedics, Yorba Linda, CA) with the following ventilatory parameters: a mean airway pressure 6 cm H2O above that required during conventional ventilation, a frequency of 15 Hz, an inspiratory fraction of 33%, and an amplitude adjusted to maintain normocapnia. These parameters were established on the basis of preliminary experiments with the aim of optimizing gas exchange in this model of lung injury (data not shown). The mean airway pressure (MAP) was then increased by 0.5- to 1-cm H2O increments until PaO2 values decreased by 20 mm Hg. Thereafter, MAP was reduced by 0.5 cm H2O until PaO2 values stabilized. These adjustments usually required approximately 30 min of the total 60-min period of HFO ventilation with blood gases analyses as performed above. The results obtained from these animals were compared with those of animals receiving only HCl and ventilated with conventional mechanical ventilation (CMV) for 60 min.
Immediately after the 60-min ventilatory period, all animals were killed by an intravascular overdose of sodium thiopental and exsanguinated via transection of the descending abdominal aorta. The chest cavity was exposed to allow for visualization of the lungs and a whole lung lavage procedure was performed as described previously (14). Briefly, approximately 30 ml/kg body weight of 0.15 M NaCl was infused via a syringe into the lungs through the endotracheal tube, which resulted in fully distended lungs. The lavage material was then gently withdrawn and reinfused twice. This procedure was repeated a total of five times and the final volume of the recovered lavage fluid was recorded. This lung lavage fluid, designated as Lavage 2, was subjected to processing within 1 h as previously described (15). Briefly, an aliquot was centrifuged at 150 × g for 10 min and the pellet containing cell debris was resuspended. The supernatant was centrifuged at 40,000 × g for 15 min at 4° C. The supernatant from the high-speed spin contained the small aggregate subfraction and the pellet containing the large aggregates was resuspended in 0.15 M NaCl. Aliquots of all the recovered fractions were subjected to lipid extraction by the chloroform–methanol method of Bligh and Dyer (16). A phosphorus assay to determine total phospholipid phosphorus was performed by a modification of the method of Duck-Chong (17) on each of the lipid extracts, including samples of Lavage 1 obtained from the appropriate animals. A Lowry protein assay (18) was performed with aliquots of Lavage 1 and Lavage 2 in order to determine the total amount of protein recovered in these samples. All phospholipid and protein values were then adjusted for body weight. The percent LA in these samples relative to the total alveolar surfactant recovered was also calculated.
Aliquots of the LA fractions obtained from each group were resuspended in 0.15 M NaCl, 1.5 mM CaCl2 to a final concentration of 1 mg of phospholipid per milliliter. The surface tension of these samples was analyzed with a pulsating bubble surfactometer as described by Enhorning (19). Briefly, a bubble is formed in a surfactant suspension to create an air–liquid interface. Ten seconds after bubble formation, the bubble is pulsated at a rate of 20 pulsations/min between a maximum radius of 0.55 mm and a minimum radius of 0.4 mm. Temperature is maintained at 37° C. The pressure across the air–liquid interface is monitored by a pressure transducer and used to calculate the surface tension at the minimum radius by the law of Young and Laplace. Values are expressed as surface tension in millinewtons per meter (mN/m) at the minimum bubble radius after 100 pulsations.
All data are expressed as means ± standard error of the mean (SEM). Comparisons among two different groups were made with an unpaired Student t test, whereas comparisons within treatment groups were made by analysis of variance (ANOVA) followed by a Tukey post-hoc test. All statistical analyses were performed with SPSS (Chicago, IL) software with probability values below 0.05 considered significant.
The average baseline PaO2 /Fi O2 value for all animals studied was 491.0 ± 3.2 mm Hg. A total of 74 animals fulfilled the inclusion criteria of PaO2 /Fi O2 values between 220 and 270 mm Hg, approximately 45 min after HCl instillation. Because of the nature of the injury, these animals represented approximately 65% of all animals that received HCl. The remainder were excluded on the basis of oxygenation values above or below the inclusion criteria. The average PaO2 /Fi O2 value at this time point was 243.4 ± 3.4 mm Hg for all animals. There were no significant differences in PaO2 /Fi O2 among any of the groups before treatment. Sample sizes of five to eight animals per group were utilized in this study.
Figure 1 shows the mean PaO2 /Fi O2 values for animals before surfactant instillation and during the 60-min ventilatory period posttreatment. Animals receiving natural ovine surfactant demonstrated significant increases in PaO2 /Fi O2 values over the first 15 min after administration when compared with pretreatment values (p < 0.01), whereas animals receiving bLES showed significant increases over the initial 10 min (p < 0.01). Both bLES and natural surfactant-treated animals had significantly higher values over the initial 15 min of ventilation compared with rSP-C-treated and control animals (p < 0.01), with the natural surfactant-treated animals remaining significantly higher than the rSP-C-treated and control groups at 30 min posttreatment (p < 0.01). Thereafter, PaO2 /Fi O2 values for bLES- and natural surfactant-treated groups steadily declined. Animals administered the rSP-C surfactant demonstrated no significant improvements in oxygenation over the 60-min period of ventilation, and were similar to the nontreated, acid-instilled control group. After 60 min of ventilation, PaO2 /Fi O2 values in all groups were similar and significantly lower than each of their respective pretreatment values (p < 0.05).
Fig. 1. Mean PaO2 /Fi O2 values after instillation of one of three surfactant preparations (100 mg of phospholipid per kilogram body weight) at time 0 (n = 7–8 animals per group) compared with non-surfactant-treated, acid controls (n = 6). Instillation of natural ovine surfactant and bLES resulted in a significant but transient increase in oxygenation, whereas rSP-C resulted in no improvements, which was similar to control animals. A progressive decline in PaO2 /Fi O2 values over the ventilatory period occurred and values were similar in all groups after 1 h. Letters represent significance (p < 0.05) versus (a) rSP-C and control, and (b) natural.
[More] [Minimize]Mean pretreatment values for PaCO2 and PIP for all animals were 53.3 ± 2.0 mm Hg and 21.5 ± 0.1 cm H2O, respectively. These values were not significantly different among the various groups at this time point. PaCO2 values generally increased after surfactant treatment, with significant differences observed over the final 30 min of ventilation compared with pretreatment values for both bLES (p < 0.01) and rSP-C-treated animals (p < 0.05) (data not shown). Nontreated, acid control animals had significantly higher PaCO2 values compared with their pretreatment values after 15 min of ventilation (p < 0.05), whereas natural surfactant-treated animals had PaCO2 values significantly higher than pretreatment values only at the 60-min time point (p < 0.01). There were no major differences among the four groups over time, with the exception of bLES-treated animals having higher PaCO2 values at 30 min compared with the natural surfactant-treated animals (p < 0.05). PIP values were significantly higher than their respective pretreatment values only for the bLES- and natural surfactant-treated animals at the 15-min time point after surfactant instillation (p < 0.05). There were no significant differences between the surfactant-treated groups over time, although all PIP values in the surfactant treatment groups were significantly higher than those of the control animals at the 15-min time point posttreatment (p < 0.05).
Table 1 shows the total protein recovery in the lung lavage obtained immediately after sacrifice (Lavage 2) for animals receiving the various surfactant preparations via instillation. Although protein recovery was extremely high in all groups (previously reported value for normal rabbits was approximately 15 mg/kg) (20), these values did not differ significantly among the groups, and were similar to the non-surfactant-treated, acid control animals. Total alveolar surfactant pool sizes recovered from the lungs of these animals at sacrifice revealed that the rSP-C- and bLES-treated groups had a significantly greater quantity of alveolar surfactant recovered compared with the natural surfactant-treated animals (p < 0.05), and non-surfactant-treated, acid control animals (p < 0.001) (Table 1). The percentage of LA recovered relative to the total surfactant pool indicated that the rSP-C-treated animals had 60.2 ± 2.5% LA, which was significantly greater than in the natural surfactant-treated group (45.1 ± 2.3% LA) (p < 0.05), but not significantly different from the bLES-treated animals (48.3 ± 3.7% LA) (p = 0.1). Furthermore, all surfactant-treated animals had a significantly greater proportion of surfactant in LA forms compared with the nontreated, acid control group (21.0 ± 2.6% LA) (p < 0.001) (Table 1). Functional analysis of the recovered LA is also shown in Table 1. Surface tension values of LA for all animals were relatively high, and did not differ significantly among the groups.
| Control (n = 6 ) | rSP-C (n = 7 ) | bLES (n = 7 ) | Natural (n = 8 ) | |||||
|---|---|---|---|---|---|---|---|---|
| Protein recovery, mg/kg | 690.8 ± 29.4 | 757.3 ± 78.0 | 737.0 ± 22.1 | 771.1 ± 74.4 | ||||
| Total phospholipid, mg/kg | 18.3 ± 0.7 | 55.8 ± 3.5† | 51.0 ± 6.9† | 32.1 ± 3.5 | ||||
| Large aggregates, % | 21.0 ± 2.6 | 60.2 ± 2.5† | 48.3 ± 3.7‡ | 45.1 ± 2.3‡ | ||||
| Surface tension, mN/m | 20.4 ± 1.6 | 20.4 ± 4.6 | 20.1 ± 1.7 | 17.0 ± 2.6 |
Figure 2 shows oxygenation values for animals undergoing the whole lung lavage procedure with the dilute surfactant suspensions. All treatment groups, including the saline-lavaged, acid-injured control group, showed similar responses. There were transient increases in PaO2 /Fi O2 values during the initial 15 min after lavage treatment, which were significantly higher than pretreatment values for all groups (p < 0.05). Natural surfactant-treated animals had PaO2 /Fi O2 values remaining significantly higher than pretreatment for a 30-min period (p < 0.05). Subsequently, there was an attenuation of this response over the remainder of the ventilatory period. PaO2 /Fi O2 values did not differ significantly among the groups at 60 min posttreatment.
Fig. 2. Mean PaO2 /Fi O2 values after surfactant lavage utilizing one of three surfactant preparations at time 0 (n = 6 animals per group) compared with saline lavage controls (n = 6). Surfactant (4-ml/kg volume, 100-mg/kg dose) was diluted in 37° C saline to a total lavage volume of 22 ml/kg. The lavage procedure resulted in a transient increase in oxygenation for all groups. A progressive deterioration in PaO2 /Fi O2 values over the ventilatory period occurred and values were similar for all groups after 1 h.
[More] [Minimize]PaCO2 values before the lavage procedure were not significantly different among the groups, averaging 52.1 ± 1.4 mm Hg. After the lavage procedure, PaCO2 values decreased and were significantly lower than pretreatment values at 5 min (p < 0.05) for all animals treated with surfactant (data not shown). PaCO2 for rSP-C-treated animals remained significantly lower than pretreatment values for 15 min (p < 0.05) after treatment. For all animals, PaCO2 values generally increased over the remainder of the ventilatory period to an average of 54.0 ± 1.3 mm Hg immediately before sacrifice. These values did not differ significantly over time among the four groups. PIP values also did not differ significantly over time between the various groups and remained relatively stable within each group over the 60 min of ventilation. The mean PIP value before treatment for all animals averaged 20.8 ± 0.4 cm H2O, and was 20.3 ± 0.5 cm H2O immediately before sacrifice for these animals.
The total protein recovered in the lavage from these animals is shown in Table 2. These results show that a large quantity of protein was removed by the initial lavage procedure (Lavage 1), although following the subsequent 60 minutes of ventilation, an additional substantial quantity of protein was present in the alveolar space (Lavage 2). Neither of these sets of values differed significantly among the various groups. Assessment of total alveolar phospholipid pools, shown in Table 2, demonstrated no significant differences between the rSP-C-, bLES-, and natural surfactant-treated animals, although these were all significantly higher than in the saline-lavaged, acid- injured control animals (p < 0.05). The percentage of total surfactant recovered as LA also did not differ significantly among the groups (Table 2). The surface activity of the surfactant measured with the pulsating bubble surfactometer demonstrated that none of the recovered LA were able to reduce surface tension to near zero values, and there were no significant differences observed among the groups (Table 2).
| Saline (n = 6 ) | rSP-C (n = 6 ) | bLES (n = 6 ) | Natural (n = 6 ) | |||||
|---|---|---|---|---|---|---|---|---|
| Protein recovery, mg/kg | ||||||||
| Lavage 1 | 176.8 ± 17.6 | 200.8 ± 14.8 | 230.6 ± 27.3 | 213.8 ± 13.7 | ||||
| Lavage 2 | 357.5 ± 44.8 | 350.6 ± 25.0 | 456.5 ± 47.9 | 381.1 ± 22.5 | ||||
| Total phospholipid, mg/kg | 9.3 ± 0.7 | 21.7 ± 1.3† | 19.4 ± 1.3† | 16.7 ± 2.2† | ||||
| Large aggregates, % | 37.4 ± 2.7 | 51.1 ± 4.0 | 46.4 ± 5.4 | 44.2 ± 4.2 | ||||
| Surface tension, mN/m | 20.7 ± 1.1 | 11.1 ± 5.3 | 19.0 ± 2.0 | 12.9 ± 5.3 |
Oxygenation responses of animals undergoing the saline lavage procedure followed immediately by surfactant instillation are shown in Figure 3. All animals demonstrated a transient increase in PaO2 /Fi O2 , with the bLES-treated animals having significantly higher values over the initial 10 min in comparison with pretreatment values (p < 0.05); however, this response diminished over the subsequent 60-min period of ventilation. There were no significant differences in PaO2 /Fi O2 values between the four groups over the ventilatory period, and all had similar values at the 60-min time point.
Fig. 3. Mean PaO2 /Fi O2 values after saline lavage and instillation of one of three surfactant preparations at time 0 (n = 5 or 6 animals per group) compared with a saline lavage control group (n = 6). Warmed saline (22 ml/kg body weight) was instilled and withdrawn followed by instillation of surfactant (100 mg/kg). This procedure resulted in a transient increase in oxygenation for all groups. A progressive deterioration in PaO2 /Fi O2 values over the ventilatory period occurred and values were similar for all groups after 1 h.
[More] [Minimize]PaCO2 values before treatment were not significantly different among any of the groups, with a mean PaCO2 value of 56.1 ± 1.7 mm Hg for all animals. The four groups showed a declining trend in PaCO2 during the initial 5 min after treatment, with the bLES-treated animals reaching significance at the 15-min sampling period, compared with pretreatment values (p < 0.05) (data not shown). Over the remainder of the 60-min ventilatory period, PaCO2 values increased in all groups and were not significantly different, averaging 58.9 ± 1.9 mm Hg immediately before sacrifice. PIP values remained relatively stable after this treatment strategy, with an average pretreatment value of 20.4 ± 0.4 cm H2O. After 15 min of ventilation, bLES-treated animals had significantly higher PIP values compared with pretreatment (p < 0.05); however, over the remainder of the ventilatory period, PIP values declined with no significant differences noted among the surfactant-treated groups. Saline-lavaged, acid-injured control animals had significantly lower PIP values compared with all surfactant-treated groups at 15 min (p < 0.05), and compared with natural surfactant-treated animals over the initial 45 min of ventilation (p < 0.05). There were no significant differences in PIP values at the 60-min time point among any of the groups, with an average value of 21.5 ± 0.5 cm H2O.
Table 3 shows the total protein recovered in both the Lavage 1 and Lavage 2 fractions of these animals. Similar to the previous treatment strategy, large amounts of protein were removed during the initial lavage procedure, although at sacrifice, a further substantial quantity of protein was recovered. There were no significant differences among the groups in the amount of protein recovered in either lavage. Analysis of total alveolar surfactant pool sizes indicated that bLES-treated animals had significantly higher pool sizes than the natural surfactant-treated animals (p < 0.05), but pool sizes were not significantly different from those of the rSP-C-treated animals. All surfactant-treated groups had significantly more alveolar surfactant than the saline-lavaged, acid-injured control animals (p < 0.001) (Table 3). Similarly, surfactant-treated animals had a significantly higher percent LA recovered than the saline-lavaged controls (p < 0.05), with an average of 60.1 ± 2.5 versus 37.4 ± 2.7% LA, respectively (Table 3). There were no significant differences in percent LA recovery between the surfactant-treated groups. In vitro analysis of the surface activity of the recovered LA did show significant differences among the groups. LA recovered from natural surfactant-treated animals had significantly lower surface tension values than the other three groups (p < 0.05), and these values were similar to those generated with the native surfactant used for treatment (data not shown). LA recovered from the rSP-C-treated animals had significantly higher surface tension values than natural surfactant-treated LA (p < 0.05), yet values were significantly lower compared with both the bLES-treated and saline-lavaged control animals (p < 0.05).
| Saline (n = 6 ) | rSP-C (n = 5 ) | bLES (n = 6 ) | Natural (n = 5 ) | |||||
|---|---|---|---|---|---|---|---|---|
| Protein recovery, mg/kg | ||||||||
| Lavage 1 | 176.8 ± 17.6 | 143.0 ± 12.8 | 167.3 ± 19.7 | 168.6 ± 8.5 | ||||
| Lavage 2 | 357.5 ± 44.8 | 321.4 ± 42.6 | 394.4 ± 47.3 | 332.0 ± 43.7 | ||||
| Total phospholipid, mg/kg | 9.3 ± 0.7† | 46.7 ± 1.7 | 56.8 ± 4.1‡ | 35.7 ± 2.3 | ||||
| Large aggregates, % | 37.4 ± 2.7† | 55.7 ± 5.2 | 60.1 ± 4.1 | 63.9 ± 5.3 | ||||
| Surface tension, mN/m | 20.7 ± 1.1 | 12.1 ± 3.2† | 22.7 ± 0.2‡,§ | 2.3 ± 0.1† |
Animals switched to HFO approximately 45 min after HCl instillation showed progressive improvements in PaO2 /Fi O2 values over time, which were significantly greater than pretreatment values, beginning at the 15-min time point (Figure 4) (p < 0.05). These sustained improvements in PaO2 /Fi O2 were significant when compared with animals undergoing conventional mechanical ventilation (CMV) over the entire 60-min period (p < 0.001). Mean pretreatment PaCO2 values did not differ significantly between the groups, and averaged 50.9 ± 2.4 mm Hg. There were no significant differences in PaCO2 values over time either within or between the groups, with an average value of 57.1 ± 4.3 mm Hg immediately before sacrifice. A similar trend was noted with the airway pressure values.
Fig. 4. Mean PaO2 /Fi O2 values after ventilation at time 0 with high-frequency oscillation (HFO) compared with conventional mechanical ventilation (CMV). HFO resulted in a significant and sustained oxygenation response over the 1-h ventilatory period compared with the CMV group. *Significant (p < 0.001) versus control.
[More] [Minimize]Total protein recovery in the HFO animals was significantly less than in animals ventilated with CMV (p < 0.001) (Table 4). Although total alveolar phospholipid pools were also significantly lower for the HFO animals (11.4 ± 0.5 mg of phospholipid per kilogram body weight) compared with CMV animals (18.3 ± 0.7 mg of phospholipid per kilogram) (p < 0.001), the percentage of LA recovered in the HFO group was significantly higher (36.8 ± 2.3 vs. 21.0 ± 2.6% LA, p < 0.01). There were no significant differences in the surface activity of the recovered LA, as all were unable to reduce surface tension to near zero values (Table 4).
| CMV (n = 6 ) | HFO (n = 6 ) | |||
|---|---|---|---|---|
| Protein recovery, mg/kg | 690.8 ± 29.4 | 363.7 ± 12.3† | ||
| Total phospholipid, mg/kg | 18.3 ± 0.7 | 11.4 ± 0.5† | ||
| Large aggregates, % | 21.0 ± 2.6 | 36.8 ± 2.3† | ||
| Surface tension, mN/m | 20.4 ± 1.6 | 17.2 ± 1.9 |
Exogenous surfactant administration is currently being investigated in patients with ARDS. Many factors have been shown to influence the efficacy of this treatment modality, which has contributed to the variability observed among the clinical trials reported to date (7, 8). It is possible that specific treatment strategies may be required for these patients on the basis of the characteristics of the patient in question. For example, a patient with a “direct insult” to the lung epithelium, such as acid aspiration, may respond to surfactant therapy in a different manner compared with a patient with an “indirect lung injury,” induced by systemic sepsis. Moreover, given the results of the current study, it is possible that for some clinical situations, exogenous surfactant may not be effective, regardless of the specific surfactant preparation and/or delivery technique utilized. An animal study showed that different types of lung injuries resulted in quite distinct alterations in the endogenous surfactant system (13). In the clinical setting, there is increasing evidence that the various risk factors leading to ARDS may involve different pathophysiological processes, although ultimately having the same physiological consequences. To obtain a better understanding of the pathophysiology of ARDS and consequently future potential treatment strategies, this study evaluated exogenous surfactant administration utilizing various preparations and delivery techniques in an animal model of acid aspiration-induced ARDS.
The lung injury model of intratracheal instillation of HCl in adult rabbits has been utilized extensively in the past as a model to reflect the clinical entity of gastric acid aspiration (13). This injury represents an immediate and direct insult to the airways and capillaries, leading to damaged epithelial and endothelial surfaces. This, in turn, results in a massive influx of proteinaceous edema fluid into the airspace (3). In the animals included in this study, the injury to the lungs was quite severe and occurred relatively early as indicated by the deterioration in oxygenation within 45 min after HCl instillation. This injury was associated with endogenous surfactant alterations including functional inactivation of the surfactant, which no doubt contributed to the progressive lung dysfunction as reflected by the marked deterioration in PaO2 /Fi O2 values over the subsequent 1 h of conventional mechanical ventilation in the nontreated, acid control group (Figure 1). The specific alterations in the endogenous surfactant system in these animals at this time point revealed low proportions of surfactant in the LA form within the airspace, which themselves were unable to adequately reduce surface tension (Table 1). This latter finding was probably due to the large amounts of protein present in the lungs of these animals. Serum proteins have been shown to inhibit surfactant function in a concentration-dependent manner (21), but can be overcome with larger concentrations of exogenous surfactant (22). In addition, the surfactant-associated protein A (SP-A) has been shown to counteract the inhibitory effects of serum proteins on surfactant function (23, 24). On the basis of these previous studies, we hypothesized that at least one of the therapeutic approaches tested in the current study would result in significant and sustained improvements in oxygenation in these animals. These approaches included the administration of relatively large quantities of exogenous surfactant, utilizing different exogenous surfactant preparations, one of which contained SP-A, and removing inhibitory serum proteins via a lung lavage procedure.
We evaluated three different surfactant preparations: (1) a natural ovine surfactant containing SP-A, SP-B, and SP-C, (2) a modified natural surfactant preparation with SP-B and SP-C, and (3) a synthetic surfactant preparation containing recombinant SP-C. These preparations were chosen on the basis of their differences in composition, as well as their previously demonstrated efficacy in lung injury models and patients with ARDS (25-27). In addition, it was reported that a low dose of rSP-C surfactant was maintained in LA forms to a greater extent than bLES when instilled into acid-injured animals (13). Because several studies have suggested that preservation of alveolar surfactant in LA forms was desirable for optimal physiological function (12, 28), we expected that animals receiving the rSP-C surfactant would have exhibited superior oxygenation responses compared with animals treated with bLES. Interestingly, there were no changes in oxygenation values immediately after the instillation of rSP-C surfactant, whereas the bLES- and natural surfactant-treated animals exhibited acute, albeit transient improvements in PaO2 /Fi O2 . The natural surfactant preparation, as predicted, resulted in superior oxygenation responses, at least initially, compared with the other two preparations in this model of severe, acute lung injury. These differences between surfactant preparations were lost, however, over the ensuing 60 min of ventilation as values for all groups were similar and significantly worse than initial values. At this time point, the severe injury elicited to the lung due to the combination of the HCl and the deleterious effects of mechanical ventilation were overwhelming, resulting in LA conversion and inactivation of the remaining surfactant existing in LA forms (Table 2). Mechanical ventilation itself has been shown to contribute to the pathophysiology of acute lung injury, particularly in lungs with a preexisting injury (28, 29). It was this combination of factors in the current study that resulted in severe and progressive lung dysfunction. Somewhat surprising, however, was the lack of response on administration of the rSP-C surfactant. A previous study by Seeger and colleagues showed that an rSP-C-based surfactant was more susceptible to protein inhibition than other surfactant preparations (21), although it has also been shown that this surfactant was physiologically equivalent to a natural surfactant when evaluated both in vitro and in vivo in a preterm rabbit model (30). Differences in protein susceptibility and physiological efficacy of various exogenous surfactant preparations are likely related to the nature of the models utilized and the methods used to assess outcomes.
In the lung injury model utilized in this study, overwhelming protein inhibition appeared to be a major contributing factor to the inferior surface tension values obtained in vitro, and consequently the suboptimal responses to exogenous surfactant instillation in vivo. We therefore evaluated alternative surfactant delivery methods involving the whole lung lavage procedure. Previous studies have shown that dilute exogenous surfactant preparations administered via whole lung lavage significantly improved the pulmonary distribution of the surfactant and resulted in superior physiological responses when compared with tracheal instillation techniques (31, 32).
On the basis of these reports, we evaluated two different therapeutic approaches, one involving a lung lavage with a surfactant suspension, and a second that consisted of lavaging the lungs with saline followed by the instillation of a large dose of surfactant. The obvious difference between these two modalities was the amount of surfactant remaining in the airspace after the procedure, although even when the quantity of surfactant remaining in the lungs was relatively large, as reflected in animals undergoing the saline lavage followed by surfactant instillation, similar results were obtained. Both strategies achieved the objective of removing large amounts of protein from the lungs, but over the subsequent 60 min of ventilation a substantial amount of protein had leaked into these severely damaged lungs. Again, it was this overwhelming pathophysiological process that likely resulted in the inhibition of surfactant within the airspace, and consequently deterioration of oxygenation values. Interestingly, analyses of the alveolar surfactant recovered at sacrifice for these animals did reveal some differences among the groups. For example, although there were similar percentages of LA recovered among the various groups, there were significant differences in the surface activity of these LA, particularly in groups treated via saline lavage followed by surfactant instillation. LA recovered from animals given the rSP-C and natural surfactant preparations had lower surface tension values compared with LA isolated from the other groups. This finding suggests that the approach of removing inhibitory proteins from the airspace in combination with leaving significant quantities of surfactant in the lung does indeed lead to more optimal functioning surfactant, at least when tested in vitro. Unfortunately, these in vitro differences did not prove to be significant in vivo, suggesting that at least for this model of acid aspiration, these animals were not “surfactant responsive.” It is possible, however, that performing multiple lung lavages and/or instilling greater quantities of surfactant into the lungs after the lavage procedure may well have proven beneficial.
To determine whether these lungs were recruitable, and that the lack of a sustained response to surfactant administration was not due to an extensively damaged, nonrecruitable lung injury, we instituted high-frequency oscillation in these animals. Previous studies have shown that HFO was able to recruit severely injured lungs and maintain this recruitment via the utilization of high distending pressures (11). In addition, because of the small tidal volumes utilized, HFO theoretically minimizes shear stress induced via phasic changes in alveolar volume associated with conventional tidal volumes (33). Furthermore, because using smaller tidal volumes has been shown to preserve surfactant in LA forms (28), HFO has the additional advantage over standard CMV approaches of minimizing surfactant aggregate conversion. Our data showed that HFO resulted in a sustained improvement in oxygenation, reduced alveolar protein leakage, and maintained surfactant in LA forms when compared with animals ventilated with CMV. Although altering the CMV strategy in these animals, such as the utilization of higher PEEP levels and lower tidal volumes, may also have improved oxygenation, comparison of various ventilatory modes was not the objective of the current study. The goal of adequate lung recruitment was effectively achieved via HFO. Interestingly, similar physiological results were obtained by Calkovska and colleagues when using HFO in a meconium aspiration model in adult rats (34). Wiswell and colleagues also found less severe histological alterations and decreased exudative debris within the airspace when HFO was utilized in piglets with lung injury induced via the intratracheal instillation of meconium (35).
In summary, we have shown that exogenous surfactant administration failed to sustain physiological improvements in oxygenation in this animal model of acid aspiration. We speculate that in this particular type of lung injury, severe permeability abnormalities resulted in protein leakage and surfactant inhibition, which mitigated prolonged alveolar recruitment. Attempts at removing some of this protein via whole lung lavage were not sufficient, likely because of the continued leakage of protein over the subsequent ventilatory period when CMV was utilized. Optimal responses were demonstrated, however, when HFO was implemented and adequate alveolar recruitment was attained. On the basis of these results, it is likely that some patients with severe aspiration-induced ARDS may not respond to exogenous surfactant treatment strategies, and that alternative methods of recruiting the lung, including the use of HFO or other protective ventilatory strategies, may be required. Future studies will address the potential added benefit of combining exogenous surfactant with HFO in these severely injured lungs.
The authors thank Dr. Fred Possmayer for the use of the pulsating bubble surfactometer.
Supported by the Medical Research Council of Canada (MRC), and the Ontario Thoracic Society. Surfactant was a generous gift by bLES Biochemicals and Byk Gulden.
| 1. | Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, Lamy M, Legall JR, Morris A, Spragg RThe American–European consensus conference on ARDS: definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med1491994818824 |
| 2. | Montgomery AB, Stager MA, Carrico CJ, Hudson EDCauses of mortality in patients with the adult respiratory distress syndrome. Am Rev Respir Dis1321985485489 |
| 3. | Greenfield LJ, Singleton RP, McCaffree DR, Coalson JJPulmonary effects of experimental graded aspiration of hydrochloric acid. Ann Surg17019697486 |
| 4. | Possmayer FA proposed nomenclature for pulmonary surfactant-associated proteins. Am Rev Respir Dis1381988990998 |
| 5. | Gregory TJ, Longmore WJ, Moxley MA, Whitsett JA, Reed CR, Fowler AA, Hudson LD, Maunder RJ, Crim C, Hyers TMSurfactant chemical composition and biophysical activity in acute respiratory distress syndrome. J Clin Invest88199119761981 |
| 6. | Veldhuizen RAW, McCaig LA, Akino T, Lewis JFPulmonary surfactant subfractions in patients with the acute respiratory distress syndrome. Am J Respir Crit Care Med152199518671871 |
| 7. | Anzueto A, Baughman RP, Guntupalli KK, Weg JG, Wiedemann HP, Raventos AA, Lemaire F, Long W, Zaccardelli DS, Pattishall ENAerosolized surfactant in adults with sepsis-induced acute respiratory distress syndrome. Exosurf Acute Respiratory Distress Syndrome Sepsis Study Group. N Engl J Med334199614171421 |
| 8. | Gregory TJ, Steinberg KP, Spragg R, Gadek JE, Hyers TM, Longmore WJ, Moxley MA, Cai GZ, Hite RD, Smith RM, Hudson LD, Crim C, Newton P, Mitchell BR, Gold AJBovine surfactant therapy for patients with acute respiratory distress syndrome. Am J Respir Crit Care Med155199713091315 |
| 9. | Sood SL, Balaraman V, Finn KC, Britton B, Uyehara CFT, Easa D1996. Exogenous surfactants in a piglet model of acute respiratory distress syndrome. Am J Respir Crit Care Med1531996820828 |
| 10. | Vanderzwan J, Veldhuizen R, Lewis JA current and future perspective of exogenous surfactant therapy for acute respiratory distress syndrome. Curr Opin Crit Care419982126 |
| 11. | Froese AB, McCulloch PR, Sugiura M, Vaclavik S, Possmayer F, Moller FOptimizing alveolar expansion prolongs the effectiveness of exogenous surfactant therapy in the adult rabbit. Am Rev Respir Dis1481993569577 |
| 12. | Ito Y, Manwell SEE, Kerr CL, Veldhuizen RAW, Yao L-J, Bjarneson D, McCaig LA, Bartlett AJ, Lewis JFEffects of ventilation strategies on the efficacy of exogenous surfactant therapy in a rabbit model of acute lung injury. Am J Respir Crit Care Med1571998149155 |
| 13. | Puligandla PS, Gill T, McCaig L, Yao L-J, Veldhuizen RAW, Possmayer F, Lewis JFThe alveolar environment influences the metabolic and biophysical properties of exogenous surfactants. J Appl Physiol88200010611071 |
| 14. | Ito Y, Goffin J, Veldhuizen R, Joseph M, Bjarneson D, McCaig L, Yao L-J, Marcou J, Lewis JTiming of exogenous surfactant administration in a rabbit model of acute lung injury. J Appl Physiol80199613571364 |
| 15. | Veldhuizen RAW, Marcou J, Yao L-J, McCaig L, Ito Y, Lewis JFAlveolar surfactant aggregate conversion in ventilated normal and injured rabbits. Am J Physiol2701996L152L158 |
| 16. | Bligh EG, Dyer WJA rapid method of total lipid extraction and purification. Can J Biochem Physiol371959911917 |
| 17. | Duck-Chong CGA rapid sensitive method for determining phospholipid phosphorus involving digestion with magnesium nitrate. Lipids141979492497 |
| 18. | Lowry OH, Rosebrough NJ, Farr AL, Randall RJProtein measurement with the Folin reagent. J Biol Chem1931951265275 |
| 19. | Enhorning GPulsating bubble technique for evaluating pulmonary surfactant. J Appl Physiol431977198203 |
| 20. | Lewis JF, Ikegami M, Jobe AHAltered surfactant function and metabolism in rabbits with acute lung injury. J Appl Physiol69199023032310 |
| 21. | Seeger W, Gunther A, Thede CDifferential sensitivity to fibrinogen inhibition of SP-C- vs. SP-B-based surfactants. Am J Physiol2611992L286L291 |
| 22. | Lachmann B, Eijking EP, So KL, Gommers DIn vivo evaluation of the inhibitory capacity of human plasma on exogenous surfactant function. Intensive Care Med201994611 |
| 23. | Cockshutt AM, Weitz J, Possmayer FPulmonary surfactant-associated protein A enhances the surface activity of lipid extract surfactant and reverses inhibition by blood proteins in vitro. Biochemistry29199084248429 |
| 24. | Sun B, Curstedt T, Lindgren G, Franzen B, Alaiya AA, Calkovska A, Robertson BBiophysical and physiological properties of a modified surfactant enriched with surfactant protein A. Eur Respir J10199719671974 |
| 25. | Rider ED, Ikegami M, Whitsett JA, Hull W, Absolom D, Jobe AHTreatment responses to surfactants containing natural surfactant proteins in preterm rabbits. Am Rev Respir Dis1471993669676 |
| 26. | Veldhuizen RAW, McCaig L, Yao L-J, Kerr CL, Ito Y, Malloy J, Lewis JFEvaluation of recombinant SP-C based surfactant in a large animal model of lung injury [abstract]. Am J Respir Crit Care Med1551997A951 |
| 27. | Lewis JF, Dhillon J, Singh RN, Johnson CC, Frewen TCExogenous surfactant therapy for pediatric patients with the acute respiratory distress syndrome. Can Respir J419972126 |
| 28. | Ito Y, Veldhuizen RAW, Yao L-J, McCaig LA, Bartlett AJ, Lewis JFVentilation strategies affect surfactant aggregate conversion in acute lung injury. Am J Respir Crit Care Med1551997493499 |
| 29. | Amato MB, Barbas CS, Medeiros DM, Magaldi RB, Schettino GP, Lorenzi-Filho G, Kairalla RA, Deheinzelin D, Munoz C, Oliveira R, Takagaki TY, Carvalho CREffect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med3381998347354 |
| 30. | Davis AJ, Jobe AH, Hafner D, Ikegami MLung function in premature lambs and rabbits treated with a recombinant SP-C surfactant. Am J Respir Crit Care Med1571998553559 |
| 31. | Balaraman V, Sood SL, Finn KC, Hashiro G, Uyehara CFT, Easa DPhysiologic response and lung distribution of lavage versus bolus Exosurf in piglets with acute lung injury. Am J Respir Crit Care Med153199618381843 |
| 32. | Gommers D, Eijking EP, So KL, van't Veen A, Lachmann BBronchoalveolar lavage with a diluted surfactant suspension prior to surfactant instillation improves the effectiveness of surfactant therapy in experimental acute respiratory distress syndrome (ARDS). Intensive Care Med241998494500 |
| 33. | Hamilton PP, Onayemi A, Smyth JA, Gillan JE, Cutz E, Froese AB, Bryan ACComparison of conventional and high-frequency ventilation: oxygenation and lung pathology. J Appl Physiol551983131138 |
| 34. | Calkovska A, Sun B, Curstedt T, Renheim G, Robertson BCombined effects of high-frequency ventilation and surfactant treatment in experimental meconium aspiration syndrome. Acta Anaesthesiol Scand431999135145 |
| 35. | Wiswell TE, Foster NH, Slayter MV, Hachey WEManagement of a piglet model of the meconium aspiration syndrome with high-frequency or conventional ventilation. Am J Dis Child146199212871293 |
