Surfactant Replacement Therapy Improves Pulmonary Mechanics in End-stage Influenza A Pneumonia in Mice 1- 3

GEERT-JAN VAN DAAL, JELLE A. H. BOS, ERIC P. EIJKING, DIEDERIK GOMMERS, EWALD HANNAPPEL, and BURKHARD LACHMANN

Introduction

Severe damage to the alveolar-capillary membrane caused by influenza A virus infection can lead to pulmonary edema, hemorrhage, and death from respiratory failure and shock (1, 2). Injury of type II cells decreases synthesis and excretion of surfactant (3). Therefore, disturbance of the pulmonary surfactant system may be an important factor in the etiology of respiratory failure during viral pneumonia, and it might be treated by surfactant replacement therapy. We recently demonstrated that surfactant instillation in rats with end -stage lethal Sendai virus pneumonia significantly improved arterial oxygenation within 5 min after instillation of surfactant (4). This effect, which lasted for at least 2 h after surfactant instillation, might be explained by improved pulmonary mechanics after surfactant instillation. To appropriately evaluate the effect of surfactant replacement on pulmonary mechanics, it is essential to measure thorax-lung compliance (Cu) and FRC. In most studies on the effect of surfactant replacement in animals with respiratory failure (for reviewsee reference 5), FRC was not determined because an effective method for measuring FRC in small laboratory animals was not available. In this review we report the effects of lethal influenza A pneumonia in mice on Ctl and on postmortem lung volume at 5 cm H 20 PEEP (V5) and the effects of surfactant replacement on Cu and V5 in end-stage influenza A pneumonia. It is speculated that V5 is the cumulative outcome of postmortem lung volume at end-expiration (supposedly approximating FRC), the added lung volume, resulting from 5 em H 20 positive pressure starting from approximately FRC, and the amount of fluid in the alveoli. V5 might therefore give a valuable impression of pulmonary stability and therefore give an impression of the status of the surfactant system and the

SUMMARY Surfactant replacement therapy may be a promising approach for treatment of respiratory failure caused by viral pneumonia. This study In mice demonstrates that during the development of lethal Influenza A pneumonia, thorax-lung compliance (Ctl/kg) and lung volume at 5 cm H20 PEEP (V5/kg) significantly decrease (28 and 54%, respectively), whereas lung water content significantly Increases (25%). Surfactant replacement therapy during the end stage of pneumonia significantly increases Ctlikg (31%) and VJkg (21%). Instillation of the vehicle for surfactant In control animals does not significantly affect Cti/kg (5% decrease), but it significantly decreases V5/kg (25% decrease). Further, a new method for postmortem measurement of lung volumes In small laboratory animals based on Archimedes' principle is presented. AM REV RESPIR DIS 1992; 145:859-863

retractive forces in the alveolar parenchyma. Further, because of the physical impossibility to measure lung volumes in mice with conventional techniques based on gas dilution, we developed a simple method based on Archimedes' principle for postmortem measurement of lung volumes in small laboratory animals; it too is described in this report. Methods Fifty-six animals were randomly divided into sevengroups (n = 8/group). One group served as a control group, and the other six groups were infected with a lethal dose of influenza A virus. In the control group and in three groups of infected animals the following protocol was executed on the first, third, and fifth day after infection. Cu, postmortem Y«, and the lung wet/dry weight ratios were measured (ratiow/d). Furthermore, on the fifth day after infection, in the three remaining infected groups, Cu was registered before and 10 min after intratracheal instillation of 0.15 ml surfactant or the vehicle for surfactant (VFS). In two groups, postmortem V5 was obtained after the second registration of Cu, and the last remaining group was used for histologic examination of the lungs.

Animals Male Swiss-bred mice 6 to 8 wk of age (SPF; Harlan/CPB, Zeist, The Netherlands) were used. Animals were housed under conventional conditions; water and food were given ab libitum. Infected animals were housed under the same conditions in a separate facility. Animals were kept under close surveillance, and body weight (BW) was registered on the day of infection and on the day of the actual measurements.

Infection Procedure Influenza virus (A/PR8/34, HINl) waspassed once in lO-dayembryonated chicken eggs.The allantoid fluid was clarified by centrifugation and stored at -70 0 C. The stock solution was diluted in sucrose and had a hemagglutination (HA) titer 1:400. Animals were exposed once for 30 min to nebulized live virus suspension (8 x dilution in PBS of stock solution) in an ll-L multichamber aerosol box (6). Aerosol with a flow of 5 L/min (particle size, 0.6 to 2.0 urn) (7) was produced with an Ultravent nebulizer (Ultravent; Malinckrodt Diagnostica, Petter, The Netherlands). Thorax-Lung Compliance For measurement of Ctl, animals were anesthetized by intraperitoneal injection of 40 mg/kg pentobarbital (Nembutal's; Algin BV, Maassluis, The Netherlands) and tracheostomized using a metal cannula as tracheal tube. Animals were paralyzed by intramuscular injection of 0.05 ml pancuronium bromide (Pavulons: Organon Technika, Boxtel, The Netherlands), transferred to amultichambered body plethysmograph (for details see

(Receivedin originalform December 13, 1990and in revised form September 18, 1991 ) 1 From the Department of Anesthesiology, Erasmus University, Rotterdam, The Netherlands, and the Department of Biochemistry, University ofErlangen, Erlangen, Germany. 2 Supported by the Dutch Foundation for Medical Research. 3 Correspondence and requests for reprints should be addressed to Burkhard Lachmann, M.D., Ph.D., Department of Anesthesiology, Erasmus University, P.O. 1738, 3000 DR Rotterdam, The Netherlands.

859

860

VAN DAAL, BOS, EIJKING, GOMMERS, HANNAPPEL, AND LACHMANN

A

reference 8) heated to 38° C, and connected to a ventilator system for pressure-controlled ventilation. The initial ventilator frequency was set at 30 pulses/min (liE ratio, 1:2; peak pressure, 25 cm H 20 [25/0]; and FIo2 = 1). The mice were ventilated under these conditions for 10 min to allow stabilization. Then Cu was recorded at a frequency of lO/min. For calculation of compliance per kilogram BW (Cu/kg), BW registered on the day of infection was used.

Lung Volume at 5 em H20 PEEP After registration of Ctl, the animals were killed with an overdose of pentobarbital and skinned. The thorax was excised, carefully leaving the diaphragm intact, and the intratracheal cannula was left in place. After this procedure the lungs were reexpanded with 0.5 ml of air to reopen atelectatic areas induced by the surgical procedure. Next, the lungs were left to collapse against a positive pressure of 5 ern H 20 and connected for 10 min to 100070 nitrogen at a pressure of 5 em H 2 0 to allow for absorption of the remaining oxygen in the lungs. The trachea was then bound, and the metal cannula was removed, maintaining the positive pressure of 5 em H 20 in the lungs. First, the total weight (W) of the thorax was registered (figure lA), and then the thorax was immersed in saline at a preset depth (figure IB) to measure the upward force (F) caused, according to Archimedes' principle, by displacement of a volume of saline equal to the volume of the thorax. V5 was calculated by the following formula (for the deduction procedure of the V5 formula, see Appendix): V5 = 0.99

*F

- 0.94

*W

For calculation of the V5 per kilogram BW (V5/kg), BW on the day of infection was used. After the V5measurement, the lungs were excised for determination of the ratiow/d. Dry lung weight was obtained after drying the lungs in a conventional microwave oven at 250 W for 1 h (9).

Surfactant/VFS Instillation On the fifth day after infection one group received treatment with surfactant and one group was sham-treated with VFS. After measurement of Cu as described above, a positive end-expiratory pressure (PEEP) of 4 to 5 cm H 20 was introduced; the rest of the ventilator settings were kept unchanged. Next, 0.15 ml of a natural bovine surfactant (200 mg/kg BW dissolved in 1:2 mixture of H 20 and saline), extracted in basically the same manner as described by others (10), or 0.15 ml VFS (1:2 mixture of H 20 and saline) was instilled into the lungs. A period of 10 min was allowed for stabilization, and Ctl was registered again. After the second registration of Ctl, V5 was measured as described above. In the last remaining group, the thorax of each animal, treated with surfactant or VFS, was opened, and a cannula was inserted into

B

MEASUREMENT OF WEIGHT

Fig. 1. Measurement of postmortem lung volume at 5 cm H20 PEEP (Vs) based on Archimedes' principle. First (A), the thorax is hooked onto the balance to measure the total weight (W) of the excised thorax. Next (B), the thorax is immersed in saline at a preset depth to measure the upward force (F)caused by displacement of a volume of saline. V5 is calculated by the following formula: V5 = 0.99 • F - 0.94 • W. (Note that because of the rigidity of the system no momentum is lost!)

MEA;;UREMENT OF UPWARD FORCE

SALINE

the right ventricle. The pressure of 5 em H 20 in the lungs was maintained while the lungs were perfused with formalin (4%) via the right ventricle. After perfusion, the lungs were removed and kept in formalin (10%) for at least 48 h. Paraffin sections werestained with hematoxylin-eosin and examined microscopically.

Statistical Analysis and Presentation of Data Statistical analysis of data was performed using the Mann-Whitney U test for analysis of intergroup differences or Wilcoxon's test for analysis of intragroup differences. Statistical significance was accepted at p ~ 0.05 (twotailed). All data are expressed as mean ± SD.

Results

From the third day after infection on, animals showed significant loss of BW (table 1) and suffered from clinically manifest illness, displaying ruffled furs, difficulties with moving, and audible breathing difficulties. Cu/kg and Vs/kg (table 1) were significantly decreased on the third and the fifth day after infection. Furthermore, ratiow/d was significantly increased on the third and fifth day after infection compared with that in healthy control animals (table 1). Surfactant instillation on the fifth day after infection (table 2) significantly increased Cu/kg (from 0.34 ± 0.15 mllcm

H 20/kg before instillation to 0.52 ± 0.28 mllcm H 2 0 / k g after instillation), whereas VFS instillation did not significantly change Cu/kg (from 0.35 ± 0.14 mllcm H 20/kg before instillation to 0.32 ± 0.15 mllcm H 20/kg after instillation). To take into account the effect of the instilled amount of fluid into the lungs 0.15 ml should be subtracted from control Vs values (or added in treated animals) to correct for the reduction of Vs by the volume of the instilled dose (assuming that no absorption of fluid has taken place). If 0.15 ml is subtracted from control Vs/kg values (20.8 ± 0.04 mIl kg), the corrected Vs/kg (V5' /kg) in control animals becomes 14.7 ± 2.2 mllkg (table 2). Evaluation of the effect of surfactant instillation on the fifth day after infection (table 2) shows that surfactant instillation almost completely restored Vs/kg (from 9.5 ± 3.3 mllkg in untreated animals on the fifth day after infection to 14.0 in surfactant-treated animals). VFS instillation significantly reduced Vs/kg (from 9.5 ± 3.3 mIlkg in untreated animals on the fifth day after infection to 4.3 ± 1.9 mIlkg in VFStreated animals) (table 2). Histologic examination of mouse lungs 5 days after infection with influenza A virus showed severe atelectasis and alve-

TABLE 1 BODY WEIGHT (BW) CHANGES, POSTMORTEM LUNG VOLUME AT 5 CM H20 PEEP (Vs!kg), THORAX-LUNG COMPLIANCE (Ctl/kg), AND LUNG WET/DRY WEIGHT RATIO (Rw/d) OF CONTROL MICE AND MICE INFECTED WITH LETHAL-DOSE INFLUENZA A VIRUS· Days after Infection Control Mice Change of BW, % RW/d Vs/kg, mllkg Cti/kg, mllkg/cm H2O

4.28 (0.51) 20.8 (2.5) 0.57 (0.08)

3 +5.5 (2.2) 4.21 (0.22) 20.0 (2.1) 0.56 (0.06)

• Values are means with SO shown in parentheses; n = a/group. p .. 0.05, Mann-Whitney U test; infected animals versus control animals.

t

-10.4 (1.5)t 4.83 (0.28)t 14.1 (2.4)t 0.46 (0.06)t

5 -21.1 (1.6)t 5.34 ·(0.78)t 9.5 (3.3)t 0.41 (0.14)t

861

SURFACTANT REPLACEMENT THERAPY IMPROVES PULMONARY MECHANICS IN INFLUENZA A PNEUMONIA

TABLE 2 INFLUENCE OF SURFACTANT OR VEHICLE FOR SURFACTANT (VFS) INSTILLATION ON THORAX·LUNG COMPLIANCE (CII/kg) AND POSTMORTEM LUNG VOLUME AT 5 em H 20 PEEP (VJkg) ON THE FIFTH DAY AFTER INFECTION WITH LETHAL·DOSE INFLUENZA A VIRUS' Surfactant

Ctl/kg, mllkg/em H 2 0t

V 5/kg, ml/kg§

VFS

Before

After

Before

After

0.34 (0.15)

0.52 (0.28):1:

0.35 (0.14)

0.32 (0.15)

Healthy

Healthy V 5'/kg

Not Treated Day 5

20.8 (0.04)

14.7 (2.2)

9.5 (3.3)

Surfactant Day 5

14.0 (2.3) II'

VFS Day 5

4.3 (1.9)

• Values are means with SO shown in parentheses.

t Gil/kg at 25 cm H20 before and 10 min after instillation of 0.15 ml surfactant or VFS.

*