Effects of Ventilation Style on Surfactant Metabolism ... - ATS Journals

Effects of Ventilation Style on Surfactant Metabolism and Treatment Response in Preterm Lambs MACHIKO IKEGAMI, KAZUKO WADA, GEORGE A. EMERSON, CELSO M. REBELLO, RAFAEL E. HERNANDEZ, and ALAN H. JOBE Department of Pediatrics, Harbor UCLA Medical Center, Torrance, California

We investigated whether the style of ventilation would influence respiratory physiology or surfactant metabolism in surfactant-treated preterm lambs. Preterm lambs were delivered at 131 6 1 d gestation and treated with an organic solvent extract of sheep surfactant (100 mg/kg). The lambs were randomized to ventilation peiods of 2 h, 5 h, 10 h, or 24 h, and to ventilation with a low rate (15 breaths/min) and high VT (15 ml/kg), with a high rate (50 breaths/min) and low VT (8 ml/kg), or with high-frequency oscillatory ventilation (HFOV). Gas exchange and lung volumes were similar across time and for the different ventilation styles. Saturated phosphatidylcholine (SatPC) in alveolar lavage was lower for the HFOV group than for the other ventilation groups at 10 h and 24 h. The rate of loss of surfactant protein B (SP-B) from these preterm animals’ lungs was slow and not influenced by ventilation style. The percentages of surfactants in large-aggregate forms were not changed by style of ventilation, and the large-aggregate surfactants had excellent function when tested in surfactantdeficient preterm rabbits. Alveolar lavage protein was low (30 ml/kg), and tissue hyaluronan did not change with time or ventilation style. In preterm lambs ventilated without causing injury, the extreme styles of ventilation examined in the study had minimal effects on lung function, surfactant function, or surfactant metabolism. Ikegami M, Wada K, Emerson GA, Rebello CM, Hernandez RE, Jobe AH. Effects of ventilation style on surfactant metabolism and treatment response in preterm lambs. AM J RESPIR CRIT CARE MED 1998;157:638–644.

Surfactant treatment of the preterm infant with lung immaturity is now routine. However, questions remain about how to best ventilate preterm infants to avoid chronic lung disease. In adult animals with lung injury, different ventilation strategies will alter both the physiologic response to surfactant treatment and the metabolism and function of surfactant (1, 2). Different ventilation styles are thought to influence the incidence of chronic lung disease in infants (3), although definitive trials of this have not been done. The physiologic response after surfactant treatment of the preterm lung depends on the type of surfactant and the use of positive end-expiratory pressure (PEEP) (4). However, there have been no studies of the relationships between ventilation styles and surfactant function or metabolism in the preterm lung. In the normal lung, large lipid arrays or large-aggregate forms of surfactant are the source of the surface film (5). This film is thought to be refined to increase the concentration of saturated phosphatidylcholine (SatPC) through changes in surface area with breathing, and small vesicles, primarily con-

(Received in original form April 1, 1997 and in revised form July 31, 1997) Supported by Grant HD 12714 from the National Institute of Child Health and Human Development. High-frequency oscillators were supplied by Sensomedics Critical Care (Yorba Linda, CA) and Bicore Monitor Systems were provided by Allied Health Care Products (Riverside, CA). Correspondence and requests for reprints should be addressed to Machiko Ikegami, M.D., Ph.D., Children’s Hospital Medical Center, 3333 Burnet Ave., Cincinnati, OH 45229-3039. Am J Respir Crit Care Med Vol 157. pp 638–644, 1998

taining lipids, reenter the hypophase for recycling or catabolism. The conversion of surfactant from large-aggregate, surface-active forms to small vesicles with poor surface activity normally maintains about 50% of the surfactant pool in forms suitable for surface film formation (6). The amount of inactive vesicular forms increases with lung injury, and this increase is associated with a deterioration in lung mechanics (7). The products of lung injury promote increased conversion of surfactant from active to inactive forms via protease enzymes (7). Surfactant isolated from the preterm lung is converted from active to inactive forms more rapidly than is surfactant from the adult when tested in vitro (8). Although SP-B is essential for normal lung function (9), there is no information available about the alveolar metabolism of SP-B in the developing lung. We therefore sought the answers to two questions, using a preterm lamb model of respiratory distress syndrome treated with surfactant: (1) what are the clearance characteristics of SP-B from preterm lung; and (2) does the style of ventilation influence surfactant metabolism and function? We selected ventilation styles at the extremes of current clinical practice to test the hypothesis that higher respiratory rates would increase and high-frequency oscillatory ventilation (HFOV) would decrease the rate of conversion of surfactant from active to inactive forms.

METHODS Surfactant Sheep surfactant was isolated from alveolar lavages of adult sheep lungs through a three-step centrifugation procedure that included a

639

Ikegami, Wada, Emerson et al.: Ventilation Style and Surfactant 0.8 M sucrose gradient (10). The surfactant was extracted three times with chloroform/methanol/saline (2:1:1 vol/vol/vol), and was resuspended with glass beads in 0.9% NaCl to a concentration of 25 mg/ml (10). This organic solvent extract of sheep surfactant contains SP-B and SP-C, but no SP-A, and is similar to surfactants used clinically.

with 0.1 mg/ml DPPC (Sigma, St. Louis, MO) in 0.9% NaCl for administration to the preterm lambs. The [125I]SP-B and aliquots of alveolar lavages from the preterm lambs were run on reduced sodium dodecylsulfate (SDS) gels, and autoradiography was used to identify the radioactivity (13).

Radiolabeled SP-B

Preterm Lambs

Sheep SP-B was isolated according to the method of Beers and colleagues (11), iodinated with Bolton–Hunter reagent (Amersham, Arlington Heights, IL), and dialyzed extensively for 48 h as previously described (13). The [125I]SP-B (0.45 mCi/ml) and [14C]palmitatelabeled dipalmitoylphosphatidylcholine (DPPC) (4.5 mCi/ml) (American Radiolabeled Chemicals, St. Louis, MO) were suspended together

Preterm lambs of 131 6 1 d gestational age were delivered by cesarean section as previously described (10, 12). Briefly, each ewe carrying twins or triplets was sedated (1 gm ketamine given intramuscularly) and given spinal-epidural anesthesia with 10 ml 2% lidocaine–0.5% marcaine (1:1 vol/vol). After delivery of the fetal head and neck, a 4.5mm-ID endotracheal tube was tied into the trachea and the fetal lung fluid that could be easily aspirated by syringe was withdrawn. Each lamb was weighed. The lipid-extracted sheep surfactant (4 ml/kg body weight) was mixed with 1 ml of the [125I]SP-B and [14C]DPPC suspension, and the mixture was instilled through the tracheal tube. To optimize surfactant distribution, no air was allowed to enter the lungs before surfactant instillation. Lambs were randomized to different ventilation styles and periods of ventilation, with ventilation initiated in the prone position. A group of lambs was ventilated with a lowrate, high-VT strategy, at a ventilatory rate of about 15 breaths/min (R15), an inspiratory time of 0.7 s, and a PEEP of 3 cm H2O. Peak inspiratory pressures (PIPs) were initially adjusted to achieve a VT of 15 ml/kg. A second group was ventilated with a high-rate, low-VT strategy, using a ventilatory rate of about 50 breaths/min (R50), an inspiratory time of 0.4 s, and a PEEP of 3 cm H2O. PIPs were initially adjusted to achieve a VT of 8 ml/kg. Conventional ventilation was conducted with pressure-limited infant ventilators (Sechrist, Anaheim, CA) delivering 16 L/min gas flow, and VT values were monitored (Bicore Monitoring Systems, Riverside, CA). For the lambs ventilated with HFOV, the first five breaths were conducted with an anesthesia bag that delivered a PIP of 35 cm H2O in order to initiate aeration and to clear surfactant from the large airways. The HFOV group was ventilated with Sensormedics 3100 ventilators (Yorba Linda, CA) set to deliver 12 Hz (720 cycles/min) using an initial mean airway pressure of 15 cm H2O. PaO2 was regulated to about 200 mm Hg by adjusting the fractional inspired oxygen concentration (FIO2), and PaCO2 was regulated to 45 to 50 mm Hg by adjusting PIP for the conventionally ventilated groups and by adjusting the amplitude for the HFOV-treated lambs. A catheter was advanced into the aorta via an umbilical artery, and was used for continuous blood pressure monitoring and for blood-gas analysis. Filtered cord blood (10 ml/kg) was given soon after birth, and a constant infusion of 5% dextrose (5 ml/kg/h) was maintained. Body temperature was maintained with radiant heat. The animals received supplemental ketamine and acepromazine every 4 h, and 2 mg/ kg gentamicin every 6 h. Functional residual capacity (FRC) was measured with the helium dilution method (14) (Equilibrated Bio Systems, Melville, NY). To estimate the lung volumes at which the lambs were oscillated, the oscillator was turned off while the airway pressure was maintained and the endotracheal tube was clamped quickly. The endotracheal tube was then opened to the helium-dilution system, and the lamb was ventilated with a conventional ventilator for about 30 s for the lung-volume measurement. Groups of five or six lambs were ventilated for periods of 2 h, 5 h, 10 h, or 24 h before they were deeply anesthetized with 25 mg/kg pentobarbital sodium 1 min after setting the FIO2 to 1.0. The tracheal tube was clamped during expiration to allow for oxygen reabsorption from the lungs.

Figure 1. Respiratory variables for all lambs ventilated with low rates (R15), high rates (R50), and high-frequency oscillatory ventilation (HFOV). The entire groups of lambs remaining alive at each time are indicated by open symbols. The animals studied at 2 h, 5 h, 10 h, and 24 h are indicated by filled symbols. (A) PaCO2 values were maintained so as not to differ among ventilation styles or durations of ventilation. (B) PaO2/FIO2 ratios were similar among the groups, except for a higher ratio for R15 than for R50 lambs at 24 h, *p , 0.05. (C) Mean airway pressures did not change over time, but were different at each time among the groups, p , 0.01. (D) Functional residual capacity (FRC) normalized to body weight did not change with time. FRC/kg was higher for the HFOV group than for the other groups at 24 h, *p , 0.05.

Pressure–Volume Curves and Lung Processing Static lung volume at 40 cm H2O pressure and the deflation limb of the pressure–volume curve were measured as described previously (12). The lungs were then removed from the thorax and thoroughly lavaged five times with saline at 48 C via the tracheal tube (8). The total alveolar lavage volume was recorded, and aliquots were used for determinations of radiolabeled SP-B, radiolabeled SatPC, SatPC content, and protein content. To isolate the large-aggregate surfactant fractions, alveolar lavage was initially centrifuged at 140 3 g for 10 min to remove debris, and was then centrifuged again at 40,000 3 g for 15 min. The pellet containing the large-aggregate surfactant was further purified by centrifugation over 0.8 M sucrose before subse-

640

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE

quent analysis (12). Homogenates of lung tissue were used for measurements of radiolabeled SP-B and SatPC and hyaluronan.

Surfactant Function Testing in Premature Rabbits Large-aggregate surfactants from the lambs and the lipid-extract sheep surfactant that was used to treat the lambs were adjusted to a concentration of 7.5 mmol SatPC/ml (about 12.5 mg lipid/ml). Surfactant function was tested in preterm rabbits of 27 d 6 2 h gestation, delivered by cesarean section (4, 12). Each rabbit received 4 ml/kg body weight of one of the surfactant samples via the trachea. The rabbits were ventilated with a FIO2 of 1.0, a rate of 30 breaths/min, and a PEEP of 3 cm H2O in a temperature-controlled ventilator–plethysmograph system maintained at 378 C. The PIPs were adjusted to give VT values of 8 ml/ kg. After 15 min ventilation, the tracheal tube was occluded for 5 min to allow absorption of oxygen. A quasistatic pressure–volume curve was generated by inflating the lungs in pressure increments of 5 cm H2O to 35 cm H2O as previously described (4). Lung volumes were recorded after 30 s at each pressure, followed by deflation of the lungs at the same pressure increments.

Materials and Measurements Lipids were extracted with chloroform–methanol (2:1, vol/vol). SatPC was recovered from lipid extracts through neutral alumina column chromatography after exposure to osmium tetroxide, and was quantified with a phosphorus assay (12). Protein was quantified according to the method of Lowry and colleagues (15). Hyaluronan was assayed with a radioassay test kit (Pharmacia, Uppsala, Sweden), using methods described by Juul and colleagues (16).

Data Analysis All values are given as means 6 SEM. Differences between groups were evaluated through analysis of variance (ANOVA), with the Student–Newman–Keuls test used for post hoc analysis.

VOL 157

1998

did the R50 lambs. FRC values were constant at about 24 ml/ kg for the R15 and R50 groups at each of the times of measurement (Figure 1D). The FRC tended to be higher for lambs ventilated with HFOV for each of the measurement times, and was significantly higher at 40 6 2 ml/kg at 24 h. Maximal lung volumes measured at 40 cm H2O from the pressure–volume curves were between 38 and 62 ml/kg at each of the study times, and there were no differences among the three styles of ventilation (Figure 2A). At 24 h the ratios of FRC to maximal lung volume were equivalent, at about 0.50, for both conventionally ventilated groups, with a ratio of 0.77 for the HFOV group (p , 0.01). Lung volumes at 10 cm H2O also were similar for all groups at all times (data not shown). Residual lung volumes at 0 cm H2O pressure were higher at 2 h and 5 h than at later times (Figure 2B). The lungs of the HFOV lambs retained less air at 0 cm H2O at 5 h than did the lungs of the R15 group. The lambs that retained significant gas volumes at 0 cm H2O had fluid visible in the trachea that prevented more complete deflation of the lungs. The lung weightto-body weight ratios decreased significantly between 5 h and 10 h (Figure 2C). There were no differences in lung weight-tobody weight ratios from 10 h to 24 h. Protein and Hyaluronan Content

The total protein in alveolar lavages was similar for all ventilation styles at all times, with an average value of about 30 mg/ kg (Table 2). There also was no effect of style of ventilation or length of ventilation on hyaluronan content expressed as mg/ kg body weight.

RESULTS Respiratory Status of Lambs

The three ventilation strategies resulted in similar PaCO2 values for the 24-h ventilation period. The entire group of lambs remaining alive at each time is shown in Figure 1A. The animals studied at 2 h, 5 h, and 10 h had mean PCO2 values similar to the values for the overall group. Oxygenation also was similar except for a small but significantly higher PaO2/FIO2 ratio for the low-ventilatory-rate (R15) animals relative to the highventilatory-rate (R50) animals at 24 h (Figure 1B). All animals had normal pH values. Ventilation in the absence of spontaneous breathing was with mean VT values of about 14 ml/kg and ventilatory pressures of about 23 cm H2O for the R15 lambs (Table 1). The R50 group had similar ventilation with VT values of about 8 to 9 ml/kg and ventilatory pressures of about 20 cm H2O. The mean airway pressures required to achieve equivalent ventilation were higher for the HFOV group than for the conventionally ventilated lambs (Figure 1C). The lambs in the R15 group had lower mean airway pressures than

TABLE 1 VENTILATION PARAMETERS FOR THE CONVENTIONAL VENTILATION GROUPS Ventilation Style Low ventilatory rate (R15) Breaths/min VT, ml/kg PIP-PEEP, cm H2O High ventilatory rate (R50) Breaths/min VT, ml/kg PIP-PEEP, cm H2O

2h

5h

10 h

24 h

15.2 6 0.7 13.8 6 0.8 21.6 6 1.2

17.1 6 0.9 14.8 6 0.7 23.0 6 1.3

17.1 6 1.1 15.1 6 0.7 25.3 6 1.7

14.7 6 0.2 12.9 6 1.1 23.0 6 1.4

50.0 6 0 8.1 6 0.9 20.2 6 1.2

49.9 6 0.1 9.5 6 1.0 20.6 6 1.1

52.0 6 2.0 9.6 6 0.8 21.2 6 1.0

50.0 6 0 7.3 6 1.2 20.4 6 0.6

Figure 2. Lung volumes from pressure–volume curves and lung weight-to-body weight ratios. (A) Maximal lung volumes at 40 cm H2O pressures for lambs that were ventilated with low rates (R15), high rates (R50), or with high-frequency oscillatory ventilation (HFOV). (B) Lung volumes measured at 0 cm H 2O lung deflation for lambs ventilated for the intervals indicated on the horizontal axis. (C) Lung weight (g)-to-body weight (kg) ratios. *p , 0.05 versus 10 h and 24 h; tp , 0.01 versus 5 h; ap , 0.01 versus R15.

641

Ikegami, Wada, Emerson et al.: Ventilation Style and Surfactant TABLE 2 PROTEIN IN ALVEOLAR LAVAGES AND HYALURONAN IN LUNG TISSUE Protein (mg/kg) Time of Ventilation Ventilation Style R15 R50 HFOV

Hyaluronan Content (mg/kg) Time of Ventilation

2h

5h

10 h

24 h

2h

5h

10 h

24 h

22.7 6 3.3 21.1 6 4.4 22.9 6 4.1

36.1 6 5.0 25.2 6 7.3 42.9 6 3.4

32.7 6 2.8 25.7 6 2.7 35.9 6 6.5

36.2 6 7.9 26.0 6 4.7 31.9 6 5.4

1.6 6 0.1 2.0 6 0.2 2.2 6 0.5

1.5 6 0.2 1.8 6 0.3 2.2 6 0.2

1.6 6 0.2 1.9 6 0.1 1.6 6 0.2

2.2 6 0.2 2.2 6 0.4 2.2 6 0.4

Surfactant Pool Sizes

The alveolar pools of SatPC for all groups decreased over the period from 2 h to 24 h, and the SatPC pools were lower for the HFOV than for the R15 group at 10 h and 24 h (Figure 3). There were no changes in tissue or total lung SatPC pool sizes with time or between groups. The percent recoveries of [14C]DPPC in lung tissue and total lung at 24 h were similar for the three ventilation styles (Table 3). There was a significant decrease in [14C]DPPC recovery in the alveolar lavages from the HFOV lambs. The percent of alveolar surfactant that was in the large-aggregate form was 91 6 1% for all groups at 2 h (Figure 4). This percent decreased to 78.1 6 2.5% by 24 h (p , 0.01 versus 2 h), and there were no significant differences between styles of ventilation. The percentages of large-aggregate forms were high in all of the groups at all times. Clearance of SP-B

The [125I]SP-B used for measurement ran as a single band, with 98% of the radiolabel recovered at about 8 kDa (Figure 5). The radiolabel was at the same molecular weight, with very little radioactivity elsewhere on the gel, when alveolar samples from lambs ventilated for 5 h were run on the gel. Therefore, the 125I radioactivity in the alveolar lavages remained associated with intact SP-B. The [125I]SP-B given at birth was lost slowly from the lungs, with recovery of about 78% of the radiolabel at 24 h (p , 0.01) (Figure 6). However, only 24% of the [125I]SP-B was recovered by alveolar lavage at 2 h, and this recovery decreased to 10% by 24 h (p , 0.01). There were no differences among styles of ventilation in the recoveries of

SP-B. Of the total [125I]SP-B in airways, 73.8 6 2.9% was recovered in large aggregate surfactant, and no differences were seen with different ventilation styles or times of ventilation. Surfactant Function

Surfactant function was tested by treating surfactant-deficient preterm rabits with either the lipid-extract natural surfactant that was used to treat the lambs or with large-aggregate surfactant recovered from the lambs by alveolar lavage. Compliance values were 0.33 6 0.01 ml/cm H2O · kg for 16 control rabbits, whereas lipid-extract, natural-sheep-surfactant-treated rabbits had compliance values of 0.70 6 0.04 ml/cm H2O · kg. Groups of five to eight rabbits were treated with the largeaggregate surfactants recovered by alveolar lavage for each style of ventilation and at each time. There were no differences in compliance after the different times of ventilation. Therefore, the data for the different times of ventilation were combined for the different ventilation styles, and were compared with the data for lipid-extracted natural sheep surfactant (Figure 7). There were small but significant increases in the compliance responses of rabbits to surfactants recovered from the R15 and HFOV groups, relative to the lipid-extracted, natural-sheep-surfactant-treated rabbits. Maximum lung volumes from pressure–volume curves were similar, although there were small but significant increases for surfactant recovered from R50 and HFOV lambs as compared with lipid-extracted sheep surfactant. The important result was that there was no decrease in the function of the large-aggregate surfactant with any of the ventilation styles over a 24 h ventilation period.

DISCUSSION We found that three styles of ventilation resulted in similar gas exchange, no indications of lung injury, and comparable surfactant metabolic and functional assessments in preterm, surfactant-treated lambs over a period of 24 h. These results are contrary to those of recent measurements of the effects of ventilation style on surfactant metabolic variables in normal and lung-injured animals (2, 17–19). The results also are contrary to those of experimental studies in preterm animals that demonstrated less injury with HFOV than with conventional ventilation (20), and to those of a recent clinical trial indicating better pulmonary outcomes for surfactant-treated preterm

TABLE 3 PERCENT RECOVERY OF [14C]DDPC AT 24 h Ventilation Style

Figure 3. Amounts of SatPC in alveolar lavages and lung tissues for lambs ventilated with low rates (R15), high rates (R50), or high-frequency oscillatory ventilation (HFOV). *p , 0.05 versus HFOV.

R15 R50 HFOV

Alveolar Lavage

Tissue

Total Lung

9.5 6 0.7 11.4 6 2.5 5.5 6 0.4*

57.2 6 5.3 53.7 6 2.5 56.8 6 3.4

66.8 6 5.8 65.1 6 1.8 62.4 6 3.3

* p , 0.05 versus R15, R50.

642


Figure 4. Percent recoveries of large-aggregate surfactant in alveolar lavages. *p , 0.01 versus 2 h.

infants with respiratory distress that were ventilated with HFOV than with conventional ventilation (3). For our study, we used previously unstressed preterm lambs at 131 d gestation because these lambs would die of severe respiratory failure within hours of delivery without surfactant treatment, and because we had previously evaluated the metabolism of the surfactant component SatPC at this gestational time (21). We treated the

Figure 5. Autoradiograph of a reduced SDS gel of [ 125I]SP-B in surfactant used to treat lambs (Lane 1) and [125I]SP-B in surfactant recovered by alveolar lavage after 5 h ventilation of three preterm lambs (Lanes 2, 3, and 4). The single bands of radiolabel are at about 8 kDa.

VOL 157

1998

Figure 6. Recoveries of [125I]SP-B from alveolar lavage, lung tissue, and total lungs of preterm lambs ventilated with low rates (R15), high rates (R50), or high-frequency oscillatory ventilation (HFOV).

lambs with an organic-solvent extract of natural sheep surfactant to mimic the types of clinically used surfactants that lack SP-A, and because this surfactant produces excellent clinical responses in preterm lambs. The HFOV strategy was to oscillate at high lung volumes, using higher MAPs than for conventional ventilation. This approach decreases lung injury in lung-injured adult animals (1, 19), and is the strategy for ventilation of the preterm human lung that has been associated with improved outcomes (3). The conventional ventilation strategies were selected to be at the extremes of standard practice for infants. We selected low

Figure 7. Surfactant function tested in preterm rabbits. Preterm rabbits were treated with the surfactant used to treat the preterm lambs (treatment surfactant) or with large-aggregate surfactants from lambs ventilated with low rates (R15), high rates (R50), or high-frequency oscillatory ventilation (HFOV). Data from groups of lambs ventilated for 2 h, 5 h, 10 h, and 24 h were combined because there were no time-dependent effects. All surfactants increased compliance and maximal lung volumes measured at 35 cm H2O pressure (V35) relative to rabbits not treated with surfactant (control), p , 0.01. Large-aggregate surfactants increased compliance and V35 above that measured with the treatment surfactant, p , 0.05.

Ikegami, Wada, Emerson et al.: Ventilation Style and Surfactant

and high ventilatory rates for lambs that were not spontaneously breathing, and adjusted PIPs to achieve normal PCO2 values. VT values were almost twice as high for the low-rate group. All three strategies resulted in similar gas exchange despite lung immaturity. Of note was that the average FIO2 at 24 h was 0.55, and that the conventionally ventilated lambs required peak PIPs minus PEEPs of 20 to 23 cm H2O. Term lambs delivered by cesarean section would require minimal oxygen and would be ventilated similarly, with only 11 cm H2O PIPs (22). Maximal lung volumes for term lambs are about 80 ml/kg, and the preterm lambs in our study had lung volumes of about 50 ml/kg. Therefore, the lambs in our study had immature lungs, and the surfactant treatment did not fully correct the abnormalities. We used total protein recovered by alveolar lavage as the indication of lung injury because it increases consistently in preterm and adult lungs with injury (17, 23). The leak of intravascular protein from the vascular space into the air space is a very sensitive indicator for the preterm lung because the latter is particularly sensitive to injury (23). The amount of protein recovered was the same (about 30 mg/kg) at all study times and for the three styles of ventilation. The amount of protein recovered by alveolar lavage of several unventilated lambs delivered at 131 d gestation was about 13 ml/kg, and was 40 mg/ kg for term lambs ventilated for 5 h (22). Therefore, the ventilation strategies used in our study produced no detectable injury on the basis of alveolar protein concentration. We also evaluated hyaluronan because this matrix component increased with the duration of ventilation and severity of lung immaturity in preterm monkeys (16). Hyaluronan also increased with oxygen exposure in newborn rats (24). The hyaluronan content of lungs from unventilated lambs was 1.5 6 0.1 mg/kg, and there was no consistent increase with duration of ventilation or ventilation style. Of note was that hyaluronan increased within 6 h in ventilated preterm monkeys (16). There is no information about changes in hyaluronan with lung injury in sheep. However, our results are consistent with minimal injury in these ventilated preterm lambs. We need to ask why so little lung injury occurred in these preterm animals’ lungs despite 24 h ventilation. One reason for this was probably a very good clinical response to surfactant treatment given at delivery and before ventilation. Treatment at birth with high-quality surfactant will result in an optimal distribution of surfactant and more uniform ventilation (14). Also, surfactant treatment at birth prevents ventilationinduced injury of the surfactant-deficient lung (20, 25). The injury that occurs with ventilation of the saline-lavaged adult lung results in part from white-blood-cell migration and activation in the lungs (26), and from ventilation of the lung at lung volumes below the normal FRC (19, 27). Surfactant treatment will not prevent the lung injury in saline-lavaged rabbits unless PEEP is also used to increase FRC (28). The preterm lung contains few white blood cells, and in unpublished observations we have not noted depletion of peripheral white blood cells in similarly ventilated lambs (29). We achieved FRC-to-maximal lung volume ratios of 0.5 and above for the three ventilation strategies because of the surfactant treatment and the selection of appropriate PEEP values. The VT of 15 ml/kg needed to ventilate lambs at a rate of 15 breaths/min is not out of the range used clinically, and the sum of FRC (about 24 ml/kg) and a VT of about 40 ml/kg was less than the maximal lung volumes of 50 to 60 ml/kg. Therefore, the low-rate–high VT strategy would not be predicted to injure the lung according to the concepts of Dreyfus and Saumon (20). We also carefully avoided hyperventilation, by VT and PCO2 monitoring. We have demonstrated that in terms of indicators of lung injury, the three

643

styles of ventilation are equivalent when a PEEP sufficient to maintain a normal FRC, and ventilation with lung volumes that do not exceed maximal lung volume, are used. In the normal adult lung, about 50% of the surfactant SatPC recovered by alveolar lavage is in a large-aggregate, surface-active form. With lung injury, the amount of largeaggregate surfactant decreases, and the function of the residual large-aggregate surfactant may be compromised (2, 7, 17). High-VT ventilation can increase the rate of loss of the largeaggregate surfactant, presumably because of large changes in lung surface area (2). There is an increased rate of loss of largeaggregate surfactant with lung injury, because of proteinases released by the injury and because of a decrease in SP-A (7). We thought that HFOV in the preterm lambs used in our study might preserve more surfactant in large-aggregate form because of the small changes in lung surface area that occur with oscillation. We also thought that the R15 group would lose more large-aggregate surfactant because of the larger changes in lung surface area with this ventilatory style. Although there was a modest decrease from 92% large-aggregate surfactant at 2 h to 73% at 24 h, and a tendency toward a lower value for the R15 group, the lambs maintained a high percentage of the surfactant in the active fraction. These results probably differ from the increased loss of large-aggregate surfactant in lung injury and for patients with ARDS, because the lambs in our study had very little lung injury. We conclude that in the minimally injured preterm lung treated with surfactant, ventilation style did not change the percentage of large-aggregate surfactant. The function of the large-aggregate surfactant also was not influenced by style of ventilation. We previously demonstrated that surfactant recovered from preterm lambs treated with several surfactants in use clinically showed improved function (11). This “activation” phenomenon occurred through the association of the clinical surfactants with the preterm lung. The surfactants used clinically in preterm infants are not as effective in terms of producing acute physiologic responses as is the organic-solvent-extracted surfactant used for the protocol in our present study (4). We anticipated that one of the ventilation styles used in our study might degrade the function of the surfactants recovered from these preterm lambs. However, surfactant recovered from the lambs showed small but significant improvements in function when tested in surfactantdeficient preterm rabbits. Therefore, the surfactant present in the lambs’ air space had good function. The different styles of ventilation used in our study could also have altered surfactant metabolism through effects on secretion. Stretching of the lung can cause surfactant release (31), and hyperventilation can cause changes in surfactantcomponent clearance (32). Froese and associates (27) showed that a ventilation strategy that optimized lung volumes in saline-lavaged and surfactant-treated rabbits resulted in better surfactant-treatment responses and better preservation of lamellar-body pools in lung tissue. In the preterm lambs in our study, alveolar SatPC pool sizes were about 20 mmol/kg at 2 h, and were similar for the three ventilation groups. The treatment dose of 100 mg/kg surfactant delivered about 60 mmol/kg SatPC to the air space. As previously observed, only about 40% of the surfactant could be recovered by alveolar lavage after surfactant treatment (21). The only differences among the treatment groups were lower recoveries of SatPC at 10 h and 24 h, and a lower recovery of [14C]DPPC at 24 h for the HFOV group. This result suggests greater lung-tissue association of surfactant with HFOV, and not the anticipated increased secretion with ventilation at higher lung volumes. SP-B is a critical component of surfactant. Surfactants deficient in SP-B have less acute effects on the surfactant-defi-

644


cient lung, and an increase in SP-B content can improve lung function (9). The lack of SP-B results in lethal respiratory distress syndrome in humans and mice (33, 34). SP-B is processed from a proprotein to the mature 7-kDa protein, which appears in lamellar bodies and is secreted in parallel with SatPC (9). SP-B is cleared from the air spaces of adult and newborn rabbits and mice at rates similar to those for SatPC, and SP-B is in part recycled back into lamellar bodies (13, 35). The biologic half-life estimates for SP-B in adult rabbit, newborn rabbit, and mouse lung were 7 h, 25 h, and 28 h, respectively. Ours is the first report of the clearance kinetics of SP-B from the preterm lung. The clearance of SP-B in the preterm is very slow and similar to previously measured values for SP-A and SatPC in the preterm lung (35). An estimate of the half-life of SP-B in the preterm lung is about 60 h, although this estimate is not precise because the half-life is longer than the last data point at 24 h. Therefore, all of the major components of surfactant that have been measured are cleared much more slowly from the preterm lung than from the adult lung. This result suggests that with recycling, the SP-B in the treatment dose of surfactant will continue to contribute to surfactant function for days. The present study demonstrates that surfactant treatment at birth can result in a minimally injured preterm lung independent of the style of ventilation. Surfactant function was preserved, and the overall metabolism of the surfactant used for treatment was not altered by the ventilation strategies used. Of note was that SP-B clearance was slow, suggesting a prolonged contribution of this protein to surfactant function. A caution about generalizing this result is that we applied the three styles of ventilation to lambs treated with a high-quality surfactant at birth, and that we avoided hyperventilation and maintained normal FRC values.

References 1. Mcculloch, P. R., P. G. Forkert, and A. B. Froese. 1988. Lung volume maintenance prevents lung injury during high frequency oscillatory ventilation in surfactant deficient rabbits. Am. Rev. Respir. Dis. 137: 1185–1192. 2. Ito, Y., R. A. W. Veldhuizen,Y. Li-Juan, L. A. McCaig, A. J. Bartlett, and J. F. Lewis. 1997. Ventilation strategies affect surfactant aggregate conversion in acute lung injury. Am. J. Respir. Crit. Care Med. 155:493–499. 3. Gertsmann, D. R., R. A. Stoddard, F. Monaco, O. Battisti, A. Francois, S. D. Minton, K. S. Meredith, J. M. Bertrand, J. P. Langhendries, and R. H. Clark. 1996. The Provo multicenter early high-frequency oscillatory ventilation trial: improved pulmonary and clinical outcome in respiratory distress syndrome. J. Pediatr. 98:1044–1056. 4. Rider, E. D., A. H. Jobe, M. Ikegami, and B. Sun. 1992. Different ventilation strategies alter surfactant responses in preterm rabbits. J. Appl. Physiol. 73:2089–2096. 5. Wright, J. R. 1990. Clearance and recycling of pulmonary surfactant. Am. J. Physiol. 259:L11–L12. 6. Gross, N. J., and K. R. Narine. 1989. Surfactant subtypes in mice: characterization and quantitation. J. Appl. Physiol. 66:342–349. 7. Higuchi, R., J. Lewis, and M. Ikegami. 1992. In vivo conversion of surfactant subtypes is altered in alveolar surfactant isolated from injured lungs. Am. Rev. Respir. Dis. 145:1416–1420. 8. Ueda, T., M. Ikegami, and A. H. Jobe. 1994. Developmental changes of sheep surfactant: in vivo function and in vitro subtype conversion. J. Appl. Physiol. 76:2701–2706. 9. Weaver, T. E., and J. A. Whitsett. 1991. Function and regulation of expression of pulmonary surfactant-associated proteins. Biochem. J. 273: 249–264. 10. Ikegami, M., C. M. Rebello, and A. H. Jobe. 1996. Surfactant inhibition by plasma: gestation age and surfactant treatment effects in preterm lambs. J. Appl. Physiol. 81:2517–2522. 11. Beers, M. F., S. R. Bates, and A. B. Fisher. 1992. Differential extraction for the rapid purification of bovine surfactant protein B. Am. J. Physiol. 6:L773–L778.

VOL 157

1998

12. Ikegami, M., T. Ueda, D. Absolom, C. Baxter, E. Rider, and A. H. Jobe. 1993. Changes in exogenous surfactant in ventilated preterm lamb lungs. Am. Rev. Respir. Dis. 148:837–844. 13. Ueda, T., M. Ikegami, M. Henry, and A. H. Jobe. 1995. Clearance of surfactant protein B from rabbit lungs. Am. J. Physiol. 268:L636–L641. 14. Ueda, T., M. Ikegami, E. D. Rider, and A. H. Jobe. 1994. Distribution of surfactant and ventilation in surfactant-treated preterm lambs. J. Appl. Physiol. 76:45–55. 15. Lowry, O. H., N. J. Rosenbrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265–275. 16. Juul, S. E., M. G. Kinsella, C. Jackson, T. A. Standaert, W. E. Truog, and W. A. Hodson. 1994. Changes in hyaluronan deposition during early respiratory distress syndrome in premature monkeys. Pediatr. Res. 35: 238–243. 17. Lewis, J. F., M. Ikegami, and A. H. Jobe. 1990. Altered surfactant function and metabolism in rabbits with acute lung injury. J. Appl. Physiol. 69:2303–2310. 18. Nicholas, T. E., J. H. T. Power, and H. A. Barr. 1990. Effect of pattern of breathing on subfractions of surfactant in tissue and alveolar compartments of the adult rat lung. Am. J. Respir. Cell Mol. Biol. 3:251–258. 19. Hamilton, P. P., A. Onayemi, J. A. Smyth, J. E. Gillan, E. Cutz, A. B. Froese, and C. Bryan. 1983. Comparison of conventional and high-frequency ventilation: oxygenation and lung pathology. J. Appl. Physiol. 55:131–138. 20. Jackson, J. C., W. E. Truog, T. A. Standaert, J. H. Murphy, S. E. Juul, E. Y. Chi, J. Hilderbandt, and W. A. Hodson. 1994. Reduction in lung injury after combined surfactant and high-frequency ventilation. Am. J. Respir. Crit. Care Med. 150:534–539. 21. Ikegami, M., A. H. Jobe, T. Yamada, A. Priestly, L. Ruffini, E. Rider, and S. Seidner. 1989. Surfactant metabolism in surfactant-treated preterm ventilated lambs. J. Appl. Physiol. 67:429–437. 22. Jobe, A. H., D. H. Polk, G. M. Ervin, J. F. Padbury, C. M. Rebello, and M. Ikegami. 1996. Preterm betamethasone treatment of fetal sheep: outcome after term delivery. J. Soc. Gynecol. Invest. 3:250–258. 23. Jobe, A. H., H. Jacobs, M. Ikegami, and D. Berry. 1985. Lung protein leaks in ventilated lambs: effects of gestational age. J. Appl. Physiol. 58:1246–1251. 24. Juul, S. E., R. C. Krueger, L. Scofield, M. B. Hershenson, and N. B. Schwarta. 1995. Hyperoxia alone causes changes in lung proteoglycans and hyaluronan in neonatal rat pups. Am. J. Respir. Cell Mol. Biol. 13:629–638. 25. Nilsson, R., G. Grossmann, and B. Robertson. 1978. Lung surfactant and the pathogenesis of neonatal bronchiolar lesions induced by artificial ventilation. Pediatr. Res. 12:249–255. 26. Kawano, T., S. Mori, M. Cybulsky, R. Burger, A. Ballin, E. Cutz, and A. C. Bryan. 1987. Effect of granulocyte depletion in a ventilated surfactant-depleted lung. J. Appl. Physiol. 62:27–33. 27. Froese, A. B., P. R. Mcculloch, M. Sugiura, S. Vaclavik, F. Possmayer, and F. Moller. 1993. Optimizing alveolar expansion prolongs the effectiveness of exogenous surfactant therapy in the adult rabbit. Am. Rev. Respir. Dis. 148:569–577. 28. Kobayashi, T., H. Kataoka, T. Ueda, S. Murakami, Y. Takada, and M. Kokubo. 1984. Effects of surfactant supplement and end-expiratory pressure in lung-lavaged rabbits. J. Appl. Physiol. 57:995–1001. 29. Carlton, D. P., J. J. Cummings, R. G. Scheerer, F. R. Poulain, and R. D. Bland. 1990. Lung overexpansion increases pulmonary microvascular protein permeability in young lambs. J. Appl. Physiol. 69:577–583. 30. Dreyfuss, D., and G. Saumon. 1993. Role of tidal volume, FRC, and endinspiratory volume in the development of pulmonary edema following mechanical ventilation. Am. Rev. Respir. Dis. 148:1194–1203. 31. Nicholas, T. E., and H. A. Barr. 1983. The release of surfactant in rat lung by brief periods of hyperventilation. Respir. Physiol. 52:69–83. 32. Oyarzun, M. J., and J. A. Clements. 1978. Control of lung surfactant by ventilation, adrenergic mediators, and prostaglandins in the rabbits. Am. Rev. Respir. Dis. 17:879–891. 33. Nogee, L. M., C. D. Demello, L. P. Dehner, and H. R. Colten. 1993. Deficiency of pulmonary surfactant protein B in congenital alveolar proteinosis. N. Engl. J. Med. 328:406–410. 34. Clark, J. C., S. E. Wert, C. J. Bachurski, M. T. Stahlman, B. R. Stripp, T. E. Weaver, and J. A. Whitsett. 1997. Targeted disruption of the surfactant protein B gene disrupts surfactant homeostasis, causing respiratory failure in newborn mice. Proc. Natl. Acad. Sci. U.S.A. 92: 7794–7798. 35. Henry, M., M. Ikegami, T. Ueda, and A. H. Jobe. 1996. Surfactant protein B metabolism in newborn rabbits. Biochim. Biophys. Acta 1300: 97–102.