Main

RDS of preterm neonates is caused by surfactant deficiency and structural immaturity of the lungs(1). Intratracheal administration of exogenous synthetic or animal-derived surfactants results in improvements of lung compliance and arterial blood gases in preterm neonates with RDS and in animal models of this disease(2, 3). Surfactants currently available for human use differ in their phospholipid composition, surfactant protein content, and mode of administration(4). Also, their surface tension characteristicsin vitro and their clinical responses vary(5, 6). This is due in part to the presence of surfactant proteins B and C in animal-derived surfactants(3). Regardless of the type of exogenous surfactant used, marked improvements in survival of neonates with RDS have been reported with this therapy(2). However, concerns about sensitization to animal proteins and the elevated cost of surfactant therapy has prompted efforts to seek additional alternatives.

We have shown previously that intratracheal administration of tensio-active agents, such as the detergent Tween 20 and several perfluorocarbons, improves lung compliance and gas exchange in surfactant-deficient animals(7–9). These agents lack the surface tension characteristics of natural surfactant and do not contain any phospholipids or surfactant proteins(8, 9). Furthermore, the perfluorocarbons tested on these studies are solid at 37°C. This is in contrast to other perfluorocarbons that are dense liquids with a relatively low surface tension (about 15-20 dynes/cm) and high solubility for oxygen and carbon dioxide. These perfluorocarbons have been used for liquid ventilation and more recently for perfluorocarbon-associated gas exchange or partial liquid ventilation(10–12).

In our previous studies using surfactant-deficient rabbits and preterm lambs, we noted marked improvements in oxygenation and lung compliance after intratracheal administration of FC-100 (3M Company, St Paul, MN) a perfluorocarbon that is solid but highly water-soluble at 37°C. These changes were better than those seen in animals treated with natural surfactant(8, 9). Furthermore, in these studies we reported that perfluorocarbons did not change alveolar surfactant content or functional residual capacity and that dynamic lung compliance correlated inversely with surface tension measurements of alveolar washes(8, 9). However, those experiments evaluated the response to a single dose of FC-100 or surfactant only over a relatively short period of time. Very immature lambs return to respiratory failure after treatment with natural surfactant(13). Repeat doses of natural surfactant produce a further, albeit diminished, improvement(13). With a longer period of mechanical ventilation after giving FC-100 than was used in our previous study, we expected severe respiratory failure to return. The present study was designed to determine whether the response to subsequent doses of FC-100 would produce a pattern of improvement such as that seen with surfactant, suggesting similarities in the pathophysiology of the response to both interventions. Hence, in the present report we used surfactant-deficient preterm lambs to evaluate the response to multiple doses of FC-100 given after meeting respiratory failure criteria, while being maintained for several hours on mechanical ventilation. Treatment with the commercially available synthetic surfactant, Exosurf Neonatal (Burroughs Wellcome, Research Triangle Park, NC), was used for comparison, because immature lambs do not survive very long if not given surfactant, and because a lack of response to Exosurf would confirm the extreme immaturity of these lambs(3, 13). This synthetic surfactant has been used recently in lambs of similar gestational age, to compare the response to surfactant given alone, or in conjunction with partial liquid ventilation(14).

METHODS

Ten date-mated ewes were anesthetized with 150 μg of Fentanyl (Janssen Pharmaceuticals, Beerse, Belgium) and underwent cesarean section after spinal anesthesia with 1% bupivacaine (Bupivan, Abbott Laboratories, North Chicago, IL) at 125 d of gestation (term = 147 ± 5 d). At this gestational age, the lambs are surfactant-deficient and rapidly develop respiratory failure(13). After the uterus was opened, arterial and venous catheters were placed in the femoral vessels of the fetus, and a tracheotomy was performed for insertion of an endotracheal tube. During this time fetal breathing was prevented by covering the snout with a warmed, normal saline-filled plastic glove. Maternal sedation throughout the procedure was maintained with repeated injections of Fentanyl while arterial blood pressure(Statham Co., Hato Rey, PR) and heart rate (Gilson ICM-5, Emeryville, CA) were monitored continuously. Also, periodic determinations of maternal and fetal arterial blood gases were done (Radiometer PHM 73, Copenhagen, Denmark).

After the preterm lambs were extracted, the umbilical cord was tied, and the fetuses were dried quickly, weighed, and placed under a radiant warmer to maintain body temperature. All lambs immediately received 0.1 mg/kg pancuronium (Pavulon, Organon Inc., West Orange, NJ) and 0.2 mg/kg midazolam(Versed, Roche Labs., Nutley, NJ) through the venous catheter. The lambs were then connected to a mechanical ventilator (Sechrist IV-100, Sechrist Industries, Anaheim, CA, or Bear Cub 2001, Bear Medical Systems, Riverside, CA) with initial settings of: rate, 40 breaths/min; PIP, 35 cm H2O; PEEP, 3 cm H2O; inspiratory time, of 0.4 s; and Fio2, 1.0. A continuous i.v. infusion of 5% dextrose at 4 mL/kg/h was given during the study period. Core body temperature was monitored continuously with a rectal thermistor and maintained at 38 ± 1°C throughout the study period. After delivery of the lambs, the ewes were killed with an overdose of sodium thiopental.

Arterial blood gases were determined at birth, at 5, 10, and 15 min after delivery, and every 15 min thereafter until the lambs showed evidence of severe respiratory failure. This was defined as a Paco2 > 8 kPa (60 torr) or Pao2 < 6.6 kPa (50 torr) despite using PIP of 35 cm H2O and Fio2 of 1.0. The lambs were then randomized to receive via the endotracheal tube 3 mL/kg 3% solution of FC-100 in normal saline(n = 5), or 5 mL/kg Exosurf reconstituted according to the manufacturer's specifications (n = 5). Both treatments were administered in two aliquots given via a side-port of the endotracheal tube adaptor over 30-60 s, followed by positioning of the lambs to either side. Arterial blood gases were measured before each dose (time 0) and at 5, 10, and 15 min after the dose of either FC-100 or Exosurf, and every 15 min thereafter. To minimize variability only PIP was modified to maintain a Paco2 < 8 kPa (60 torr) with a maximum PIP of 40 cm H2O. Determinations of tidal volume, PIP, and PEEP were performed before administration of any dose of FC-100 or Exosurf, and every 15 min afterward. Tidal volume was obtained by measuring gas flow at the proximal end of the endotracheal tube with a Fleisch No. 0 pneumotachograph attached to a differential pressure transducer (Validyne MP45-14-871, Northridge, CA) and an electronic flow integrator (Validyne MC 1-3). These measurements were registered using a Gilson ICM-5 multichannel recorder. Dynamic compliance was calculated directly from these readings as the volume difference between points of no flow (end-inspiratory volume minus end-expiratory volume) divided by the pressure difference at those points. Values were then normalized to body weight. The VEI was calculated as follows(15):Equation

Second and third doses of FC-100 or Exosurf were given for evidence of respiratory failure as outlined above. We planned to give these only after at least approximately 90-120 min had elapsed since the previous dose. Throughout the entire study, arterial blood pressure and heart rate of the lambs were monitored continuously using a Statham pressure transducer connected to a Gilson ICM-5 multichannel polygraph. At the end of the study period (6-7 h), the lambs were killed with an overdose of sodium thiopental.

Statistical comparisons between physiologic data at time 0 and subsequent values for each dose were done using ANOVA for repeated measures with Newman-Keuls post hoc testing for multiple comparisons. Comparisons between the FC-100- and Exosurf-treated lambs at each time point were performed using two-way ANOVA. Survival was compared using the Mantel-Haenszel test(16). A p < 0.05 was considered significant. All values are reported as means ± SEM, except for survival, which is reported as percentage of the total. The study protocol was approved by the Institutional Animal Review Committee of the participating institutions.

RESULTS

There were no differences in birth weight, sex distribution, or fetal arterial blood gases between the two groups of lambs (Table 1). Likewise, the time from birth until the criteria for diagnosis of respiratory failure was met, and the postnatal age when either FC-100 or Exosurf were given, was comparable between both groups. In all lambs, severe respiratory failure was evident before 1 h after delivery.

Table 1 Characteristics of the preterm lambs
Full size table

Administration of the initial dose of FC-100 resulted in a rapid rise in Pao2, and a fall in Paco2 (Fig. 1). In contrast, Exosurf administration resulted in no significant changes. An increase of arterial pH, which reflected the changes of Paco2, was observed in the five lambs given FC-100 (data not shown). The increase in Pao2 seen with FC-100 lasted between 30 and 45 min in most lambs, although due to the relatively small number of lambs studied. Pao2 values were significantly higher than those at time 0 for only 15 min after the initial dose. However, Pao2 and Paco2 values observed after FC-100 remained significantly better than those of Exosurf-treated lambs throughout the 90-min period. Second and third doses of FC-100 were given to all lambs of that group, but only three lambs from the Exosurf group received a third dose of this surfactant, because the other lambs had died. Furthermore, the third dose of Exosurf was given before 90 min from the second dose in one lamb, to avoid having less than three lambs to evaluate. Nonetheless, there were no differences between these groups in the age at which the doses of FC-100 or Exosurf were given (Table 1). Pao2 increased again after the second dose of FC-100 compared with values at time 0 for this dose; however, the changes observed were of a lesser magnitude than those noted with the initial administration of this compound (Fig. 1). As seen with the initial dose of FC-100, Pao2 and Paco2 remained significantly better than in Exosurf-treated lambs for most of this second 90-min period. No significant increases of Pao2 occurred with the third dose of FC-100. Soon after the repeat doses of FC-100 were given, Paco2 increased transiently, and arterial pH decreased accordingly. No significant improvements of Pao2 were noted after the repeat doses of Exosurf. Moreover, Paco2 rose progressively with a concomitant fall in arterial pH in these lambs.

Figure 1
figure 1

Effects of repeated doses of FC-100 (♦) or Exosurf(○) on (A) Pao2, and (B) Paco2. Time 0 represents measurements before each dose. *p < 0.05vs Exosurf group, †p < 0.05 vs time 0 for that dose in same group.

Full size image

Dynamic lung compliance improved after the initial dose of FC-100 and remained significantly higher than that of the Exosurf-treated lambs(Fig. 2), whose lung compliance remained unchanged throughout the study. Also, the VEI improved markedly in lambs treated with FC-100 (Fig. 2). The improvements of dynamic lung compliance and VEI were sustained throughout the study, although there were transient decreases of VEI soon after the second and third doses of FC-100. In Exosurf-treated lambs, VEI decreased despite the repeated administration of this surfactant.

Figure 2
figure 2

Effects of repeated doses of FC-100 or Exosurf on (A) dynamic lung compliance, and (B) ventilatory efficiency index (VEI). Symbols as in Fig 1. *p < 0.05vs Exosurf group, †p < 0.05 vs time 0 for that dose in the same group.

Full size image

All lambs treated with FC-100 survived for at least 6 h after delivery, whereas only one Exosurf-treated lamb was alive at that time(Fig. 3, p < 0.05). All deaths in the Exosurf group appeared to be due to severe acidosis and respiratory failure, and no pneumothoraces occurred in either group. Mean arterial blood pressure and heart rate slowly, but progressively, decreased after treatment with FC-100, whereas values in the Exosurf-group were unchanged(Fig. 4). Arterial blood pressure also fell in surviving lambs from this group toward the end of the study period, although no changes of heart rate were noted.

Figure 3
figure 3

Survival of preterm lambs treated with FC-100 or Exosurf. Symbols as in Fig 1. *p < 0.05 by Mantel-Haenszel test.

Full size image
Figure 4
figure 4

Changes in mean arterial blood pressure (MABP, lower tracings) and heart rate in beats/min (bpm, upper tracings) in preterm lambs treated with FC-100 or Exosurf. Symbols as inFig 1. *p < 0.05 vs Exosurf group.

Full size image

DISCUSSION

Surfactant deficiency results in poor lung compliance and impaired gas exchange. Both of these abnormalities improve after intratracheal administration of surfactants with biophysical properties similar to those of natural surfactants(1, 17, 18). These abnormalities of lung function can also be improved using tensio-active agents with biophysical properties different from pulmonary surfactant(7–9). In surfactant-deficient animals, we have shown that intratracheal administration of a single dose of FC-100, given at birth or after respiratory failure ensues, results in more marked increases in oxygenation and dynamic lung compliance than those observed with natural surfactant(9). Those studies, however, were of relatively short duration (30-60 min) and did not assess the response to additional doses of FC-100.

This study confirms that instillation of FC-100 into the airway improves lung function in surfactant-deficient lambs with severe respiratory failure. In addition, these data show that when poor oxygenation reappears despite the initial treatment with FC-100, further improvements in oxygenation can occur with repeat dosing of this perfluorocarbon. The increase in Pao2 seen after the intial dose of FC-100 in this study was less pronounced than that reported previously, which is probably a reflection of the higher mean airway pressure used to ventilate the lambs in our previous studies, and possibly, of the lower dose of FC-100 used in the present study(7, 9). Similarly, other investigators have noted more marked improvements of Pao2 after administration of progressively higher amounts of Perflubron(Alliance Pharmaceutical Corp., San Diego, CA) to rabbits with surfactant deficiency and to infants with RDS(19, 20). Perflubron has a surface tension of about 18 dynes/cm, which is comparable to that of FC-100 (15 dynes/cm)(10). However, a major difference between perfluorocarbons such as Perflubron and FC-100 is that the latter spreads rapidly at an air/water interface. This allows small doses of FC-100 to function like surfactant. Perfluorocarbons used for liquid ventilation would not function in this manner. The rapid spreading of FC-100 is consistent with the very even distribution of this agent compared with surfactant found in our previous study(9). Recently, the use of Perflubron in a group of preterm infants with severe respiratory distress syndrome was reported(20). In this study Perflubron was administered in larger volumes than in our lamb studies, using the technique called partial liquid ventilation. After instillation of Perflubron into the lungs of these infants, there were significant improvements of oxygenation and lung compliance. Administration of large volumes of Perflubron into the lungs results in alveolar recruitment, lowered surface tension, and better distribution of pulmonary blood flow.

After the second dose of FC-100, Pao2 increased less than with the initial dose and remained significantly higher than in Exosurf-treated lambs. Similar results have been found after giving repeat doses of exogenous surfactant to preterm neonates and lambs(13, 21). Although not measured, this is presumably due to barotrauma that is observed in the immature lungs of these lambs after a period of mechanical ventilation(9). The recurrence of poor oxygenation cannot be explained by a loss of this perfluorocarbon by evaporation from the lungs, because FC-100, unlike the perfluorocarbons used for partial or total liquid ventilation, is a solid at body temperature and does not have a measurable vapor pressure. There are no studies of the metabolism of FC-100 in the lungs.

In contrast to the lambs treated with FC-100, the lambs treated with Exosurf did not show increases in oxygenation even after receiving multiple doses of this synthetic surfactant. This finding is in agreement with that of Cummings et al.(3) and Leach et al.(14) who studied immature lambs of similar gestational age. However, more mature lambs do exhibit improved oxygenation after treatment with Exosurf(22, 23). This confirms the severe lung immaturity of these lambs and implies that FC-100 might be more effective in improving oxygenation than synthetic surfactant in immature subjects. Another reason for the poor response to Exosurf may be lack of surfactant proteins B and C, because the addition of a mixture of these two proteins to this surfactant can improve its ability to inflate excised rat lungs in vitro(24). We administered additional doses of Exosurf to surviving lambs within a shorter interval than that used clinically(25, 26). This decision was made primarily as an attempt to match the repeated treatments with either compound under investigation. No deleterious effects after using a higher phospholipid dose have been found in animals or humans(4, 27, 28).

Although the increases in oxygenation with FC-100 usually lasted less than 1 h, improvements of Paco2 and arterial pH were more persistent. Also, the VEI, which takes into account the ventilatory pressures and rates necessary to maintain a certain Paco2, was higher over a more prolonged period of time, although it decreased transiently shortly after the repeat doses of FC-100. Note, however that further improvements in VEI or compliance were not obtained with repeated doses of FC-100. Conversely, lambs treated with Exosurf did not exhibit any improvements in Paco2, pH, or VEI, but rather a worsening trend in these variables. These findings are also similar to those of Cummings et al.(3) and Leachet al.(14). In our FC-100-treated lambs, dynamic lung compliance improved after the first dose of this compound and remained significantly higher than in Exosurf-treated lambs for the remainder of the study. FC-100 and several other perfluorocarbons studied previously by us and other investigators are capable of increasing dynamic lung compliance of surfactant-deficient subjects or those with lungs injured through various mechanisms(8, 9, 12, 20, 29, 30). However, they have very little effect in normal lungs with adequate surfactant pools(9). Moreover, the improved dynamic lung complaince observed after administration of FC-100 is not associated with a higher functional residual capacity(9). Despite this persistently improved dynamic lung compliance after giving FC-100, Pao2 fell after an initial increase. This fall in Pao2 was probably due to worsening ventilation/perfusion abnormalities.

We do not have a clear explanation for the slow but progressive fall in mean arterial blood pressure and heart rate noted in lambs treated with FC-100. Previous studies using FC-100 done in lambs of the same gestational age did not reveal any cardiovascular changes(9), although those experiments were much shorter than our current protocol. The changes reported here were not accompanied by metabolic acidosis or obvious decreases in tissue perfusion. Rather, they were associated initially with an improving pH and may reflect an improved metabolic state of the lambs through the first few hours after delivery. Perfluorocarbons are generally inert and only large i.v. doses of some of these compounds can produce tissue damage(31), thus perfluorocarbon toxicity seems to be an unlikely explanation for our findings.

In summary, we have shown that repeated intratracheal administration of FC-100 improved gas exchange and lung function in surfactant-deficient lambs for many hours without obvious side effects. These improvements were more marked than those observed after treatment with the commercially available synthetic surfactant Exosurf. These findings along with our previous reports(8, 9), suggest that agents with tensio-active properties comparable to FC-100 might improve lung function of infants with RDS. However, further laboratory studies are needed to elucidate other aspects such as the metabolic fate of perfluorocarbons such as FC-100 in the lungs, and the long-term safety of their use.

Acknowledgments. The authors thank Carlos Muñoz, Hernán Riquelme, and Bernardo Barrales for their excellent technical assistance, and Leticia Molina for her help in preparation of this manuscript.