Aerosolized surfactant JAMES

F. LEWIS,

MACHIKO

treatment IKEGAMI,

ALAN

of preterm lambs H. JOBE,

Department of Pediatrics, Division of Neonutology, Harbor-UCLA Torrance, California 90509

LEWIS, JAMES F., MACHIKO IKEGAMI, ALAN H. JOBE,AND BANNIE TABOR. Aerosolized surfactant treatment of preterm lambs. J. Appl. Physiol. 70(Z): 869-876,1991.-To evaluate the potential for aerosolized surfactant treatments of surfactant deficiency, twin lamb fetuses were delivered at 130-132 days gestational ageand received nebulized natural surfactant (Neb NS), nebulized Survanta (Neb Surv), tracheally instilled natural surfactant (Inst NS), or nebulized saline (Neb Saline). Neb NS and Neb Surv groups had significant increasesin ventilatory efficiency index and dynamic compliance values (P < 0.05). Both groups also had pressure-volumecurves that were comparableto the Inst NS group. The Neb Saline control group had deterioration of the ventilation efficiency index and dynamic compliancevalues over time aswell aspressure-volume curves that demonstratedsmaller lung volumes comparedwith all three surfactant-treated groups (P < 0.01). Delivery of aerosolized surfactant to the lung was only -2 mg lipid/kg for the nebulized groups,a doseone-twentieth of that previously noted to be effective in instillation protocols. Distribution histograms of the aerosolizedsurfactant-treated groups differed from the instilled animals as there was more deposition in the right upper lobesand tracheae in the nebulized groupscompared with the instilled group (P < 0.05). Pulmonary blood flow was not altered by aerosolizedsurfactant treatment. Administration of aerosolizedsurfactant to preterm lambs improved lung function at a very low surfactant dose. nebulization; respiratory distress syndrome; surfactant distribution; phosphatidylcholine

SURFACTANT REPLACEMENT is an effective therapy

for infant respiratory distress syndrome (RDS) because it lowers the incidence of barotrauma, death, and bronchopulmonary dysplasia (18, 24). Presently, whether given at birth to “prevent” RDS or as a “rescue” treatment for established RDS (7,24), the method of administration is to disconnect the infant from the ventilator and instill aliquots of the material through the endotracheal tube with positioning of the infant in an attempt to optimize distribution. The distribution pattern of instilled surfactant is relatively uniform when given at birth to the fluidfilled lungs of preterm lambs; however, this is not the case if the lungs have been ventilated before treatment (20,34). Although it was recently shown that the homogeneity of surfactant distribution in aerated rabbit lungs was directly related to the volume administered (8), instillation of excess fluid to injured lungs may not be desirable. Despite the fact that a more uniform distribution pattern of tracheally instilled surfactant resulted in a superior clinical response (20,34) and that aerosolized parti0161-7567191

$1.50

AND

BANNIE

TABOR

Medical Center,

cles were more uniformly distributed than tracheally instilled particles in a direct comparison (3), a major concern of aerosolizing surfactant is that enough material be delivered at the alveolar level to elicit a physiological response. Although the initial clinical trials of surfactant replacement used aerosolized dipalmitoylphosphatidylcholine (DPPC) (4,31), the results were inconclusive, and since then there have been very few reports of surfactant administration using this method of delivery. Given the advances in knowledge of surfactant treatments, as well as the availability of technically superior nebulizers, we tested aerosolization of natural surfactant in a surfactant-deficient lamb model of prematurity that predictably responds to tracheally instilled surfactant (17). We compared both the physiological response and quantity of material deposited in lung tissue using both modes of surfactant delivery. We also compared the effects of aerosolized natural surfactant with a modified bovine lung surfactant, Survanta, that is presently being used clinically. METHODS Surfactant preparation. Natural surfactant was isolated from alveolar washes of adult sheep or cows as previously described (19). An aliquot of the natural surfactant was radiolabeled with [3H]choline-labeled DPPC in the form of liposomes that were associated with the natural surfactant and recovered by centrifugation at 27,000 g for 15 min at 4OC (13). This pellet was diluted with 0.45% NaCl, and an aliquot was taken to determine radioactivity. The rest of the isolated natural surfactant was kept frozen at -2OOC. [3H]choline-labeled Survanta (10 &i/ml) was prepared by Ross Laboratories. Both labeled and unlabeled Survanta were supplied at a concentration of 25 mg lipid/ml. The two surfactant solutions for nebulization were prepared by mixing 5-&i aliquots of the labeled natural surfactant and Survanta with their respective unlabeled surfactants and diluting these solutions with 0.45% NaCl to a final concentration of 20 mg lipid/ml. Aliquots of these final solutions were taken in duplicate as reference samples for recovery data. Animals given surfactant by tracheal instillation received 50 mg lipid/kg body weight of natural surfactant containing 1 &i of the 3H-labeled natural surfactant in a volume of 4 ml/kg body wt (10). Delivery and ventilation of the lambs. Date-mated Western mixed breed ewes carrying twins or triplets at l30132 days gestation age (term 150 days) were preanesthetized with 1 g ketamine and 2.4 mg atropine given by

Copyright 0 1991 the American Physiological

Society

869

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.

870

AEROSOLIZED

SURFACTANT

intramuscular injection. Spinal-epidural anesthesia was administered using a 1:l (vol/vol) mixture of 2% lidoCaine and 0.5% marcaine. The head and neck of each lamb were delivered through a midline uterine incision as previously described (19). Briefly, the fetal trachea was exposed and an endotracheal tube (4.5 mm ID) was secured by tracheostomy. Approximately 5-10 ml of fetal lung fluid were removed, and the endotracheal tube was clamped. A 3.5Fr infant feeding tube was passed into the superior vena cava via the external jugular vein using 1% lidocaine as local anesthesia and then secured. After sampling of-cord venous blood for pH and blood gas measurements, the lamb was delivered, weighed, and placed on a time-cycled pressure-limited Sechrist infant ventilator (Sechrist Instruments, Anaheim, CA). Ventilator settings were 100% oxygen at a flow of 10 l/min, an initial peak inspiratory pressure (PIP) of 35 cmH20, a positive end-expiratory pressure (PEEP) of 3 cmH,O, a frequency (f) of 30 breaths/min, and an inspiratory time of 1 s. Subsequently, the PIP and f were changed in an attempt to normalize PCO~ values between 30 and 40 Torr. Peak inspiratory pressures higher than 40 cmH,O were not used to avoid pneumothoraces. A 5-Fr catheter was placed in the distal aorta via an umbilical artery for monitoring of blood pressure, heart rate, and blood gas measurements. Lambs were ventilated in a sternumdown position, and anesthesia was maintained throughout ventilation to prevent spontaneous respiration with intermittent intramuscular injections of 10 mg/kg ketamine and 0.1 mg/kg acepromazine. Body temperature was maintained between 37 and 39OC with radiant warmers and warming pads, and each lamb received a continuous infusion of 100 mg kg-‘. 24 h-l of 5% dextrose via the arterial catheter. The animals also received lo-ml/kg transfusions of filtered maternal blood for hypotension and/or blood replacement. Treatment protocol. All lambs were stabilized on the ventilator before treatment to assess the degree of lung immaturity in each lamb before any treatment was given. Each lamb was randomized to receive nebulized natural surfactant, nebulized Survanta, tracheally instilled natural surfactant, or nebulized saline. For the nebulized groups, treatments were initiated 30 min after the start of ventilation and continued until the animals were killed, while the instilled group was treated after 2 h of ventilation. The instilled animals were treated by disconnecting them from the ventilator, slowly injecting the surfactant solution at room temperature through the endotracheal tube while the animal was rotated, and then reconnecting them to the ventilator for the duration of the experiment (19). The possible effects of the treatments on pulmonary blood flow distributions were measured using radiolabeled microspheres (9). 46Sc- and s5Sr-labeled microspheres with 15 -t 2 pm (SD) diam (3M Medical Products, St. Paul, MN) were mixed with 5 ml of maternal blood and injected over 30 s into the superior vena cava. The 46Sc microspheres were given just before initiation of any of the treatments and the %Sr microspheres were given 0.5 h before the animals were killed. At the end of the study period, the tracheal tube was clamped for 5 min to permit absorption atelectasis. All l

TREATMENT

animals were killed 3.5 h after birth with an overdose of pentobarbital sodium followed by exsanguination. Nebulization procedure. A Vortran Intermittent Signal Actuated Nebulizer (VISAN, Vortran Medical Techology, Sacramento, CA) was connected to the ventilator circuit just proximal to the endotracheal tube (29). The nebulizer assembly consisted of a Vortran disposable compressed-air-operated nebulizer with a 5-ml reservoir. This was connected to a T piece, which was in turn connected to the ventilator tubing at one end and the endotracheal tube at the other. A short piece of tubing from the expiratory circuit was connected to a filter to trap aerosolized radioactive surfactant within the system. At 30 min of age, 5 ml of the labeled natural surfactant, Survanta, or unlabeled normal saline were added to the reservoir, which was placed in ice, and nebulization was initiated. A solenoid valve activated the nebulizer only during the inspiratory phase of the breathing cycle using a timing signal from the ventilator (29). Because the nebulizer was designed for adult patient use, a flow of 100% oxygen at 10 l/min was required to produce optimal particle sizes. Surfactant particle size at this flow rate was measured for the aerosolized natural surfactant and Survanta using an ARIES multijet seven-stage cascade impactor (ARIES, Davis, CA). The mass median aerodynamic diameters of particles generated for both compounds were 1.8-2.3 pm (k2.1 geometric SD), which are ideal size distributions for deep lung deposition (28). Then 5-ml aliquots of the solutions were added intermittently to the reservoir so that aerosolization was maintained throughout the remainder of ventilation and total volume of solution added to the reservoir during the experiment was recorded. After the animals were killed, the circuit was disconnected, the reservoir was rinsed with methanol, and lipids were extracted (2). Lipid extracts in chloroform were dried under nitrogen, and aliquots were used to determine the total amount of residual label left in the reservoir. Similar procedures were performed on the T piece and expiratory tubing of the nebulizer assembly. Physiological measurements. Blood gases and pH were monitored every 15-30 min throughout the experiment, and the ventilation efficiency index (VEI) was subsequently calculated according to Notter et al. (26). This measurement evaluates the overall ventilation efficiency of mechanically ventilated animals to compensate for the differences resulting from the combined effects of changing ventilatory pressures, rates, and PCO~ values. It is calculated from the following equation: 3,800 + (AP . f Pa,,,), where 3,800 is a CO, production constant (ml. Torr . kg-’ . min-‘), AP is PIP - PEEP (cmH,O), f is respiratory frequency, and Pace is arterial PCO, (Torr). Tidal volumes were measure% with a pneumotachometer and dynamic compliance was calculated as tidal volume divided by PIP - PEEP and by body weight in kilograms. Compliance measurements were obtained immediately before treatment was initiated and 30 min before the animals were killed. Immediately after each animal was killed, quasi-static pressure-volume curves were measured from 0 to 40 cmH,O at 5-cmH,O pressure increments after 30 s at each pressure. Processing of lungs. Immediately after pressure-voll

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.

AEROSOLIZED

TABLE

1. Description of lambs

SURFACTANT

’

RESULTS Pretreatment

Nebulized Natural Survanta Saline

animals surfactant

Instilled animals Natural surfactant Values

are means

871

TREATMENT

n

Birth Weight, kg

Ventilation Efficiency Index

Dynamic Compliance, ml kg-’ - cmH,O-’

11 7 7

2.7k0.2 2.3kO.l 2.5t0.3

0.08~0.01 O.lOtO.O1 0.09~0.01

0.29t0.03 0.37kO.03 0.34t0.05

4

2.6kO.3

0.09t0.02

0.23kO.05

l

k SE.

ume curve measurements, the lungs of the lambs were removed intact with the endotracheal tube in place. They were then inflated with air to 35 cmH,O pressure, and the lobes were separated with small pieces of aluminum foil (20). The lungs were frozen by immersion in liquid nitrogen and stored at -2OOC. Each lung was then cut into a mean of 103 t 4 pieces with a band saw, and lobar location of each piece and weight were recorded. The microsphere-associated radioactivity was determined for all pieces, and each piece was then homogenized in 3 ml of saline using an Ultra-turax homogenizer (Tekmar, Cincinnati, OH). Aliquots were taken for protein determination according to Lowry et al. (22) as well as for lipid extraction with chloroform-methanol (2). Lipid extracts were dried under nitrogen, and 3H-label radioactivity was determined by liquid scintillation counting. Aliquots of the solutions used to treat each lung also were extracted and counted. Data analysis and presentation. To determine the amount of radiolabel actually reaching lung tissue in the nebulized surfactant groups, the total radioactivity of material added to the reservoir was calculated from the volume of solution added. Residual radioactivity remaining in the nebulizer reservoir was subtracted to determine the amount actually nebulized into the system. The sum of radioactivity of all lung pieces was divided by the total amount nebulized to determine the percent of nebulized surfactant deposited in the lung tissue. Distribution of blood flow pre- and posttreatment as well as surfactant distributions were calculated by dividing the amount of radioactivity per milligram protein for each lung piece by the mean value of radioactivity per milligram protein for all pieces to normalize the numbers. These normalized values are presented as histograms with interval widths of 10% about the mean value of 1.0. All pieces having a normalized value co.15 or >2.0 times the mean were grouped at the extremes of the distribution intervals (20). All values are presented as means & SE, and differences in means between more than two groups were tested for significance by analysis of variance followed by the student Newman-Keuls multiple comparison procedure. Pretreatment and maximum posttreatment values for PO,, VEI, and dynamic compliances were compared using paired t tests.

Description of animals, Cord pH and blood gases were similar for all animals at delivery (data not shown). The number of lambs, birth weights, and mean pretreatment values of VEI and the dynamic compliances for each group are shown in Table 1. Although there was variability among animals in the initial ventilatory period before treatment, the mean values for the VEIs and dynamic compliances for the four groups were similar. Animals given nebulized saline represented a control group to document any physiological effects that occurred due to the nebulization procedure itself. Animals treated with tracheally instilled natural surfactant served as a comparison group for the nebulized animals to document treatment responses and surfactant distribution patterns. Physiological responses.Maximal VEI values were significantly higher than pretreatment values (P < 0.01) for animals given nebulized natural surfactant or Survanta, whereas sequential values for the nebulized saline animals were all lower than pretreatment values (Fig. 1). VEI improved after surfactant was given by tracheal instillation to values similar to those found in the nebulized surfactant-treated groups. The mean values for PO, before treatment and throughout the time of ventilation after treatment for each group are shown in Fig. 2. The PO, values initially tended to increase in the nebulized surfactant-treated groups and were maintained at reasonable levels over the study period for animals given nebulized Survanta. The PO, values for the nebulized saline groups were lower than the nebulized surfactant-treated groups after treata-ii v -c

.-w 6 >

I-

0.20

A-

I

0.00 9 PRE-R,

I 30

@ Neb.NS W Neb.Surv. A Neb.Saline

1

I

I

I

I

60

90

120

150

180

TIME AFTER TREATMENT(min) Inst.NS 0.15

!/t-t

0.10 I

1 --+--+ 11

1

I

/I

I

l

-*-

1

1

0.05 t

1 0.00

4

30

I

I

60

90

TIME AFlER FIG. 1. Curves for ventilation ment (Pre-Rx) and throughout animals given nebulized natural vanta (Neb Surv), or nebulized SE. Values for animals given (Inst NS) are shown throughout surfactant administration 120

1

120

1

150

.

180

1

210

BIRTH(min)

efficiency index (VEI) before treattime of ventilation after treatment for surfactant (Neb NS), nebulized Sursaline (Neb Saline). Values are means k tracheally instilled natural surfactant time of ventilation after birth with min after start of ventilation (t).

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.

AEROSOLIZED

872

a-0 m-m A -A

SURFACTANT

TREATMENT

401

Neb.NS Neb.Surv. Neb.Soline

NebSurv. Inst.NS Neb.NS

04 PRE-R,

1 3o

I 60

I 90

TIME AtlER

I 120

I 150

Neb.Soline

I 180

TREATMENT(min)

0

0

Inst.NS

1

5

10

15

20

25

30

35

40

PRESSURE (cm H20> 4. Quasi-static pressure-volume curves of preterm lamb lungs given nebulized natural surfactant (Neb NS), nebulized Survanta (Neb Surv), nebulized saline (Neb Saline), or instilled natural surfactant (In& NS). Values at maximal pressure and all values on deflation limbs are significantly higher for surfactant-treated groups compared with saline-treated group (P < 0.01). FIG.

of animals

---A

1

animals

1 60

0 30

r 90

1 120

TIME Al7ER

I 150

1 180

I 210

BlRTH(min)

2. Curves for PO, (Torr). before treatment (Pre-Rx) and throughout time of ventilation after treatment for animals given nebulized natural surfactant (Neb NS), nebulized Survanta (Neb Surv), or nebulized saline (Neb Saline). Values are means t SE. Values for animals given tracheally instilled natural surfactant (Inst NS) are shown throughout time of ventilation after birth with surfactant administraFIG.

tion

120 min after

start

of ventilation

(f ).

ment, and they progressively decreased with time. The instilled group had stable PO, values until surfactant instillation at 2 h after birth, and the PO, values then increased significantly after surfactant treatment (P
2.0 times mean value was significantly higher for the Neb NS and Neb Surv groups vs. Inst NS group (P < 0.05). FIG.

those animals given surfactant as an aerosol compared with the instilled group, the only significant difference between the groups was that there was a greater percentage of pieces with more than two times the mean value for the nebulized natural surfactant group (12.9 -t 0.6%) and nebulized Survanta (11.4 t 2.1%) vs. the instilled surfactant group (7.6 t 0.6%) (P < 0.05). We then assessed where this deposition was occurring within the lung by comparing the lobar distributions between the groups (Fig. 6). There was significantly more recovery in the tracheae for the aerosolized groups compared with the instilled group (P < 0.05). Within the lung

0 t.15

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0 j2.0

DISTRIBUTION INTERVALS Normalized distributions of the ratio (pretreatment ‘Qz/ posttreatment %Sr) of injected microspheres for lambs treated with nebulized Survanta (Neb Surv). All values are calculated as described in METHODS and expressed as mean percentage of pieces of lung in each 10% distribution interval. FIG.

7.

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.

874

AEROSOLIZED

SURFACTANT

TREATMENT

parison group demonstrating surfactant treatment responsiveness for the nebulized surfactant-treated aniSurfactant replacement is currently an accepted mals. We felt that although in&tilling surfactant at the method of treatment for infant RDS (7, 18, 24) and is same early time point that nebulization was started is being investigated as a therapeutic modality for the adult clinically more realistic, it would not be a fair comparison RDS (ARDS) (21, 30, 35). In both these clinical situaof the two methods of delivery, as it would take a period tions, the optimal dose, timing, and method of surfactant of time for aerosolized surfactant to accumulate in lung delivery require further study. Animal models are useful tissue. The specific time of 2 h after birth that the train this regard, not only to help understand the clinical cheally instilled surfactant was administered was thereimpact of different treatment strategies but also to per- fore chosen as it represented the midpoint of nebulizamit metabolic and functional studies not possible in hu- tion and would allow us to compare the physiological remans (8, 17, 18, 33); For example, the intratracheal in- sponse and distribution patterns of the two methods of stillation of surfactant before the first breath in premadelivery after a period of mechanical ventilation. Ventiture lambs resulted in a superior clinical response and a lation itself is known to increase protein leak and cause more uniform surfactant distribution than did surfactant lung damage, and both effects will tend to decrease the instillation given after a short period of ventilation (20, magnitude of the surfactant treatment response at in34). Practically, surfactant cannot always be given at creasing times after birth in preterm lambs (19,25). Venbirth and multiple dosing may be required, and there tilation also causes increased secretion of endogenous remains the question as to the optimal mode of adminissurfactant with most of this response occurring within tration of this material to an aerated lung. A nonhomothe first 20 min after initiating ventilation (16). Nebuligeneous distribution of surfactant could create areas zation was therefore started 30 min after ventilation so with different compliances within the lung and result in that most of these endogenous metabolic changes had overdistension of some segments. occurred and also to allow stabilization of the animals If a more uniform distribution of exogenous surfactant with documentation of lung immaturity before treatcontributes to an improved clinical response, then aero- ment. solization of the material would seem to be a superior Gas exchange was significantly improved by treatment method of delivery. Brain et al. (3) showed that the pulwith both nebulized natural surfactant and Survanta. monary deposition of submicronic iron oxide particles PO, values for the nebulized Survanta group increased when given by aerosol inhalation had a more uniform and were maintained reasonably well throughout the distribution pattern, resulted in a deeper penetration of treatment period. Although this improvement in oxygenlung tissue, and had fewer areas with very little deposiation was not statistically significant, PO, values are tion compared with similar particles given by the traknown to be quite variable in immature animals and cheal instillation technique. This method of delivery also some of this variability is likely related to patent ductus seems more physiological and technically more convearteriosus and/or atria1 shunts (5). Therefore, despite nient than giving volumes of fluid via the tracheal tube oxygenation being the most common criteria of surfacwith rotation of the immature infant. Although the initant response in humans, compliance and PCO, retial clinical trials of surfactant treatment for RDS did in sponses have been utilized in preterm lambs (17). There fact use aerosolized DPPC, the results of these trials were small differences in the maximal responses and the were either inconclusive (31) or did not lead to clearly duration of responses in VEI values for the two different “beneficial effects” (4). The main reason a greater effect aerosolized surfactants. For example, the animals was not seen was felt to be that the material used treated with nebulized Survanta tended to have a more (DPPC) did not have optimal surface properties rather rapid improvement in VEI than animals treated with nethan as a result of the problems with the method of delivbulized natural surfactant. This was in contrast to a ery. Subsequently, there have been a few conflicting re- more rapid response to tracheally instilled natural surports of the effects of aerosolized surfactant on pressurefactant than Survanta when they were given to similar volume curves of excised surfactant-deficient animal gestational age lambs at birth (14). Different responses lungs (12,32), and there was a recent report of administo different surfactants could have a number of explanatering an aerosolized artificial surfactant to oleic acid-intions such as selective inactivation of the surfactants by jured sheep (35). Although there was no documented proteins in these immature lungs (11) or different metabeneficial effect acutely, the aerosolized surfactant treatbolic fates when delivered as an aerosol, although we ment appeared to protect against a repeated lung injury doubt that nebulized surfactant would behave much difin those animals ferently metabolically than instilled surfactant. Despite the variabilities noted in gas exchange, there is We decided to pursue this method of treatment in view of the potential benefits of a superior distribution patconvincing evidence of a positive response in lung mechanics as reflected by the compliances and pressure-votern of aerosolized surfactant particles in aerated lungs lume relationships. The increases in dynamic compliusing technologically advanced nebulization equipment ance may have been even more significant had we mea(29). RDS of the preterm lamb is an extensively studied model that is similar physiologically and pathologically sured them earlier as the posttreatment values were taken just before the animals were killed so that radioacto the premature infant (17,33). We studied a nebulized saline control group to document any effects from the tive aerosolization was not interrupted. The compliance nebulization itself. The instilled surfactant group was values probably were not maximal values based on the not meant to represent a control group but merely a comVEI measurements. DISCUSSION

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.

AEROSOLIZED

SURFACTANT

In contrast, animals given nebulized saline all had significant deterioration of oxygenation, VEI, and dynamic compliance values over time as well as less volume accumulation on pressure-volume curves at the time they were killed. This control group is important as it verifies that the physiological improvements noted in the nebulized surfactant-treated groups was a result of the surfactant delivered to the lung rather than the aerosol assembly itself. The most surprising finding in this study was the small quantity of surfactant necessary to produce these physiological changes when-that surfactant was delivered as an . aerosol. Previous studies have indicated that with instillation at birth a minimum of 19 mg lipid/kg natural surfactant was required to improve gas exchange and more surfactant was required to improve pressure-volume curves in 120-day-preterm lambs (10). Consequently, an empiric dose of 50 mg/kg has been accepted as the minimal amount of surfactant required to effectively treat these animals. The amount of aerosolized material actually reaching lung tissue in this study was -2 mg lipid/kg for both natural surfactant and Survanta. This is lower than the endogenous surfactant pool size of 3-4 mg/kg that is required to increase compliance values in preterm rabbits (15) and is comparable to the theoretical dose of 3 mg/kg that would be necessary to form a monomolecular film on the alveolar surface (23). The high percentage of recovery of label in the instilled group verified minimal loss of material through processing. Also, these recovery estimates in the lung tissue should not be complicated by catabolic activity as we previously documented the virtual lack of catabolism of exogenously administered surfactant in preterm lambs at this gestational age (14). Our initial hypothesis to explain this response with such a low dose delivered to the lungs was that it was due to a superior distribution of surfactant when given as an aerosol. However, we found no significant difference in the percentage of pieces of lung within &25% of the mean value for animals given nebulized natural surfactant (24.0 t 2.4%), nebulized Survanta (25.4 -t 5.6%), or instilled surfactant (26.3 t 1.0%). The greater recovery of material in the right upper lobe in the aerosolized animals was similar to the findings of Brain et al. (3) using similar methods with iron oxide particles. If the right upper lobe were omitted from our data, and histograms were recalculated for all three treatment groups, the two nebulized surfactant groups had improved distribution patterns in the remaining lung, as the percentage of pieces of lung within &25% of the mean increased by -8% for each of the two nebulized surfactant groups. The instilled group, on the other hand, had a similar distribution pattern when the right upper lobe was omitted and no difference in the percentage of pieces within &25% of the mean. Despite the improvement in distribution of surfactant in the aerosolized animals when the right upper lobe was omitted, there were still no striking differences in distribution between the three groups. Our interpretation of these findings is that our methods of assessing distribution of surfactant are of low resolution relative to surfactant distribution at the alveolar level. Our original hypothesis, therefore, was not adequately tested given the

TREATMENT

875

information available. Nevertheless, one can assume that the aerosolized surfactant was deposited at the critical locations in the distal lung to achieve the physiological effects noted. The reason for the increased delivery of surfactant and iron oxide particles to the right upper lobe is unknown but may be due to preferential ventilation of this lobe. Although we did get a significant response to the low surfactant dose delivered to the lungs, the efficiency of delivery was such that the total amount of surfactant required was much greater than by tracheal instillation. The minute ventilation was 0.6-0.7 l/min for these animals, and the flow from the nebulizer was 10 l/min. Because the nebulizer was triggered on inspiration only, it delivered -5 l/min of aerosol to the system. Therefore, a maximum of only 12-S% of the flow from the nebulizer containing the surfactant was going into the lungs, and only a fraction of that would be expected to settle rather than be exhaled. The amount of material actually deposited in lung tissue was -15% of the aerosol that entered the lung, a number that was somewhat higher than what was quoted in the literature for drug delivery by aerosolization (6). However, it is difficult to compare different aerosols directly because of the different nebulizers used, droplet sizes, drug solubilities, and different absorption of the materials to tubing. Nevertheless, with improvements in design to decrease flow and thus decrease the losses through the expiratory circuit, this method of delivery may prove to be more efficient than the present tracheal instillation technique. Further studies of surfactant delivery as an aerosol are warranted to document its clinical role as a primary and/ or secondary treatment modality for RDS. It may also be an ideal mode of administering surfactant to the aerated damaged lungs of ARDS patients who theoretically would require large volumes of surfactant if given intratracheally. The superior distribution pattern of low dose surfactant would also make this an ideal “carrier” of other drugs administered simultaneously (1,27). We thank Dr. Darryl Absolom, Ross Laboratories, for support and encouragement. This work was supported by National Institute of Child Health and Human Development Grants HD-12714 and HD-11932, Canadian Medical Research Council Fellowship 33703 (J. F. Lewis), and a grant from Ross Laboratories. The nebulizer was kindly provided by Vortran (Sacramento, CA). Address for reprint requests: J. F. Lewis, Harbor-UCLA Medical Center Dept. of Pediatrics, Div. of Perinatal/Neonatal Medicine, 1000 W. Carson St., Bldg. A-17 Annex, Torrance, CA 90509 Received 21 May 1990; accepted in final form 15 September 1990. REFERENCES

1. ABRA, R. M., C. A. HUNT, AND D. T. LAU. Liposome disposition in vivo VI: delivery to the lung. J. Pharmacol. Sci. 73: 203-206, 1984. 2. BLIGH, E. F., AND W. J. DYER. A rapid method of total lipid extraction and purification. Can. J. Biochem. 37: 911-917, 1959. 3. BRAIN, J. D., D. E. KNUDSON, S. P. SOROKIN, AND M. A. DAVIS. Pulmonary distribution of particles given by intratracheal instillation or by aerosol inhalation. Enuiron. Res. 11: 13-33, 1976. 4. CHU, J., J. A. CLEMENTS, E. K. COLTON, M. H. KLAUS, A. Y. SWEET, AND W. H. TOOLEY. Neonatal pulmonary ischemia: clinical and physiologic studies. Pediatrics 40: 709-782, 1967. 5. CLYMAN, R. I., A. JOBE, M. HEYMANN, M. IKEGAMI, C. ROMAN, B. PAYNE, AND F. MAURAY. Increased shunt through the patent duc-

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.

AEROSOLIZED

876

SURFACTANT

tus arteriosus after surfactant replacement therapy. J. Pediatr. 1: 101-107,1982. 6. FLAVIN, M., O’BRODOVICH.

22.

7.

23.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

M. MACDONALD, M. DOLOVICH, G. COATES, AND H. Aerosol delivery to the lung with an infant ventilator. Pediatr. Pulmonol. 2: 36-39, 1986. FUJISVARA, T., S. CHIDA, Y. WATABE, H. MAETA, T. MORITA, AND T. ABE. Artificial surfactant therapy in hyaline-membrane disease. Lancet 1: 55-59, 1980. GILLLARD, N., P. RICHMOND, T. A. MEFWIT, AND R. G. SPFUGG. Effect of volume and dose on the pulmonary distribution of exogenous surfactant administered to normal rabbits or to rabbits with oleic acid lung injury. Am. Rev. Respir. Dis. 141: 743-747, 1990. HEYMANN, M. A., B. D. PAYNE, J. I. E. HOFFMAN, AND A. M. RuDOLPH. Blood flow measurements with radionuclide-labeled particles, Prog. Cardiovasc. Dis. 1: 55-79, 1977. IKEGAMI, M., F. H. ADAMS, B. TOWERS, AND A. B. OSHER. The quantity of natural surfactant necessary to prevent the respiratory distress syndrome in premature lambs. Pediatr. Res. 14: 1082-1085, 1980. IKEGAMI, M., Y. AGATA, T. ELKADY, M. HALLMAN, D. BERRY, AND A. JOBE. Comparison of four surfactants: In vitro surface properties and responses of preterm lambs to treatment at birth. Pediatrics 79: 38-46, 1987. IKEGAMI, M., T. HESTERBERG, M. NOZAKI, AND F. H. ADAMS. Restoration of lung pressure-volume characteristics with surfactant: comparison of nebulization versus instillation and natural versus synthetic surfactant. Pediatr. Res. 11: 178-182, 1977. IKEGAMI, M., A. JOBE, AND G. DUANE. Liposomes of dipalmitoylphosphatidylcholine associate with natural surfactant. Biochim. Biophys. Acta 835: 352-359,1985. IKEGAMI, M., A. JOBE, T. YAMADA, A. PRIESTLY, L. RUFFINI, E. RIDER, AND S. SEIDNER. Surfactant metabolism in surfactanttreated preterm ventilated lambs. J. Appl. Physiol. 67: 429-437, 1989. IKEGAMI, M., A. H. JOBE, T. YAMADA, AND S. SEIDNER. Relationship between saturated phosphatidylcholine pool sizes and compliance of preterm rabbit lungs. Am. Rev. Respir. Dis. 139: 367-369, 1989. JACOBS, H., A. JOBE, M. IKEGAMI, AND S. JONES. Accumulation of alveolar surfactant following delivery and ventilation of premature lambs. Exp. Lung Res. 8: 125~140,1985. JOBE, A., AND M. IKEGAMI. The prematurely delivered lamb as a model for studies of neonatal adaptation. In: Animal Models in Fetal Medicine, edited by P. W. Nathanielsz. Ithaca, NY: Perinatology, 1984, p. l-30. JOBE, A., AND M. IKEGAMI. Surfactant for the treatment of respiratory distress syndrome. Am. Rev. Respir. Dis. 136: 1256-1275,1987. JOBE, A., M. IKEGAMI, T. GLATZ, Y. YOSHIDA, E. DIAKOMANOLIS, AND J. PADBURY. Duration and characteristics of treatment of premature lambs with natural surfactant. J. Clin. Invest. 67: 370-375, 1981. JOBE, A., M. IKEGAMI, H. JACOBS, AND S. JONES. Surfactant and pulmonary blood flow distributions following treatment of premature lambs with natural surfactant. J. Clin. Invest. 73: 848-856, 1974. LACHMANN, B. Surfactant replacement in acute respiratory failure: animal studies and first clinical trials. In: Surfactant Replace-

24.

25.

26.

27

28.

29.

30.

31.

32.

33.

34.

35.

TREATMENT

ment Therapy in Neonatal and Adult R.D.S., edited by B. Lachmann. Berlin: Springer-Verlag, 1987, p. 212-223. LOWRY 0. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275,1951. MARKS, L. B., R. H. NOTTER,G. OBERDORSTER, AND J. T. MCBRIDE. Ultrasonic and jet aerosolization of phospholipids and the effects on surface activity. Pediatr. Res. 17: 742-747, 1983. MERRIIT, T. A., M. HALLMAN, B. T. BLOOM, C. BERRY, K. BENIRSCHKE, D. SAHN, T. KEY, D. EDWARDS, A. JARVENPAA, M. POHJAVOIRI, K. KANKAANPAA, M. KUNNAS, H. PAATERO, J. RAPOLA, AND J. JAASKELAINEN. Prophylactic treatment of very premature infants with human surfactant. N. Engl. J. Med. 315: 785790,1986. NILSSON, R., G. GROSSMAN, AND B. ROBERTSON. Lung surfactant and the pathogenesis of neonatal bronchiolar lesions induced by artificial ventilation. Pediatr. Res. 12: 249-255, 1978. NOTI’ER, R. H., E. A. EGAN, M. S. KWONG, B. A. HOLM, AND D. L. SHAPIRO. Lung surfactant replacement in premature lambs with extracted lipids from bovine lung lavage: effects of dose, dispersion technique, and gestational age. Pediatr. Res. 19: 569-577, 1985. PETTENAZZO, A., A. JOBE, M. IKEGAMI, R. ABRA, E. HOGUE, AND P. MIHALKO. Clearance of phosphatidylcholine and cholesterol from liposomes, liposomes loaded with metaproterenol, and rabbit surfactant from adult rabbit lungs. Am. Rev. Respir. Dis. 139: 752-758, 1989. RAABE, 0. G. Deposition and cIearance of inhaled particles. In: Occupational Lung Diseases, edited by J. B. L. Gee, W. K. C. Morgan, and S. M. Brooks. New York: Raven, 1984, p. l-38. RAABE, 0. G., J. I. C. LEE, AND G. A. WONG. A signal actuated nebulizer for use with breathing machines. J. Aerosp. Med. 2: 201210,1989. RICHMAN, P. S., SPRAGG, R. G., B. ROBERTSON, T. A. MERRITT, AND T. CURSTEDT. The adult respiratory distress syndrome: first trials with surfactant replacement. Eur. Respir. J. 2, Suppl. 3: 10% 111,1989. ROBILLARD, E., Y. ALARIE, P. DAGENAIS-PERUSSE, E. BARIL, AND A. GUILBEAULT. Microaerosol administration of synthetic @r-d& palmitoyl-L-a-lecithin in the respiratory distress syndrome: a preliminary report. Can. Med. Assoc. J. 90: 55-57, 1964. SHANNON, D. C., H. KA~EMI, E. W. MERRILL, K. A. SMITH, AND P. S. L. WONG. Restoration of volume-pressure curves with a lecithin fog. J. Appl. Physiol. 28: 470-473,1969. STAHLMAN, M., V. S. LEQUIRE, W. C. YOUNG, R. T. BIRMINGHAM, G. A. PAYNE, AND J. GRAY. Pathophysiology of respiratory distress in newborn lambs. Am. J. Dis. Child. 108: 375-392,1964. WALTHER, F. J., I. M. KUIPERS, G. E. M. GIDDING, D. WILLEBRAND, R. T. F. BUCHHOLTZ, AND E. M. BEVER~. A comparison of high-frequency oscillation superimposed onto backup mechanical ventilation and conventional mechanical ventilation on the distribution of exogenous surfactant in premature Iambs. Pediatr. Res. 22: 725-729, 1987. ZELTER, M., B. J. ESCUDIER, J. M. HOEFFEZ, AND J. F. MURRAY. Effects of aerosolized artificial surfactant on repeated oleic acid injury in sheep. Am. Rev. Respir. Dis. 141: 1014-1019,199O.

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.